#pip install pingouin

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.preprocessing import StandardScaler
from scipy.stats import zscore
import pingouin as pg
import statsmodels.api as sm
from statsmodels.stats.outliers_influence import variance_inflation_factor
from statsmodels.formula.api import mixedlm
import warnings
import os

warnings.filterwarnings('ignore')
pd.set_option('display.max_rows', 50)
pd.set_option('display.max_columns', 500)

os.chdir('C:/Users/Ryo/OneDrive/Desktop/Master Thesis/master_thesis/study3')

data preparation

prep for the response df

file_loc = 'raw/Study3 - final_July_18_2024_02_35_final.xlsx'
df = pd.read_excel(file_loc)
df

	ResponseId	Ads_OP_1_Reversed	Ads_OP_2_Reversed	Ads_OP_3_Reversed	Ads_OP_4_Reversed	Ads_OP_5_Reversed	Ads_OP_6_Reversed	Ads_CO_1_Normal	Ads_CO_2_Normal	Ads_CO_3_Normal	Ads_CO_4_Normal	Ads_CO_5_Normal	Ads_CO_6_Normal	Ads_EX_1_Reversed	Ads_EX_2_Reversed	Ads_EX_3_Reversed	Ads_EX_4_Reversed	Ads_EX_5_Reversed	Ads_EX_6_Reversed	Ads_AG_1_Normal	Ads_AG_2_Normal	Ads_AG_3_Normal	Ads_AG_4_Normal	Ads_AG_5_Normal	Ads_AG_6_Normal	Desc_OP_1_Normal	Desc_CO_1_Reversed	Desc_EX_1_Normal	Desc_AG_1_Reversed
0	5db4f0b63e33f2000dd54016	22	19	14	16	17	15	21	22	18	18	18	15	24	25	23	22	21	17	70	69	69	68	65	61	75	26	75	22
1	66294a585f1cf3fdaeb80120	28	50	47	72	50	41	34	50	50	59	46	21	46	50	58	62	55	38	56	50	55	64	58	60	91	90	80	40
2	665a52573b9527ce0011158b	10	10	10	10	10	10	0	0	0	0	0	0	10	10	10	10	10	10	61	60	61	61	61	62	100	100	100	42
3	651ebcca48c60acc82f1f2bd	60	50	63	90	90	89	31	50	34	50	39	37	62	50	65	80	70	79	62	50	54	67	67	77	82	82	82	34
4	6634fb892af2e227ab65f004	80	50	90	85	90	85	80	50	75	50	75	87	95	50	95	50	95	95	70	60	70	67	70	70	10	80	60	80
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
87	64d52f62a1f1a7371760fe4f	82	64	86	72	79	90	40	44	24	37	31	15	39	31	21	21	13	13	81	73	91	68	81	89	20	70	83	32
88	66622736b3fcb6b8f3c04fdd	88	86	92	83	86	85	28	41	11	28	28	16	81	78	87	81	75	86	68	66	70	61	70	66	76	26	68	30
89	6660680acd027329cecdf2b8	37	33	35	49	47	46	61	56	67	53	74	86	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
90	5d215a1bbf7f840019701939	53	52	70	62	75	82	35	49	28	40	31	33	69	69	77	64	69	68	53	54	55	51	50	50	51	52	53	53
91	663294f21bab6d7c3f7bf27b	26	46	29	42	56	34	40	40	18	37	37	46	6	17	8	8	5	3	58	87	46	84	55	54	69	54	93	46

92 rows × 29 columns

print(f"shape {df.shape}")

shape (92, 29)

file_loc = 'raw/20240721_prolific_export.csv'
target_respondants = pd.read_csv(file_loc)
target_respondants = target_respondants[target_respondants['Status'] == 'APPROVED']['Participant id'].unique().tolist()
len(target_respondants)

90

study3_respondants = df['ResponseId'].tolist()
print(f"original respondse counts: {len(study3_respondants)}")

# filter df to target participants
# Count non-null values for each row
df['non_null_count'] = df.notna().sum(axis=1)

# Sort by 'ResponseId' and 'non_null_count' (descending), then drop duplicates
df = df.sort_values(['ResponseId', 'non_null_count'], ascending=[True, False]).drop_duplicates('ResponseId')

# Remove the temporary 'non_null_count' column
df = df.drop('non_null_count', axis=1)

# Filter to keep only target respondents
s3_response_df = df[df['ResponseId'].isin(target_respondants)]
print(f"final respondse counts: {len(s3_response_df['ResponseId'])}")

original respondse counts: 92
final respondse counts: 90

# Reverse scores for columns ending with '_Reversed'
reversed_columns = [col for col in s3_response_df.columns if col.endswith('_Reversed')]
for col in reversed_columns:
    s3_response_df[col] = 100 - s3_response_df[col]

# Remove '_Normal' or '_Reversed' from all column names
s3_response_df.columns = s3_response_df.columns.str.replace('_Normal', '').str.replace('_Reversed', '')

# Display the modified DataFrame
s3_response_df

	ResponseId	Ads_OP_1	Ads_OP_2	Ads_OP_3	Ads_OP_4	Ads_OP_5	Ads_OP_6	Ads_CO_1	Ads_CO_2	Ads_CO_3	Ads_CO_4	Ads_CO_5	Ads_CO_6	Ads_EX_1	Ads_EX_2	Ads_EX_3	Ads_EX_4	Ads_EX_5	Ads_EX_6	Ads_AG_1	Ads_AG_2	Ads_AG_3	Ads_AG_4	Ads_AG_5	Ads_AG_6	Desc_OP_1	Desc_CO_1	Desc_EX_1	Desc_AG_1
12	577f70d1454e5400015f1845	72	59	70	34	88	83	88	85	91	87	95	93	7	29	36	5	20	10	82	89	91	88	91	91	91	85	88	6
81	5a2adf6a8e00a000019864fb	72	55	64	67	58	70	27	39	35	35	22	29	74	70	82	77	75	80	80	76	76	66	77	67	72	25	26	71
33	5aa806e7777df200016088c5	20	9	0	0	0	0	16	0	6	0	11	0	16	0	2	10	4	0	62	69	54	38	48	67	84	74	23	59
63	5afe053059ae1e00017e6a3c	78	19	17	16	17	19	0	20	18	62	9	10	47	46	47	27	46	47	44	44	47	47	47	47	83	72	70	64
10	5b757a03976ecb0001b1168f	8	19	0	0	0	0	64	73	73	65	68	70	27	39	14	8	17	7	71	84	77	68	92	93	45	27	81	42
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
82	66645cff5b7cb46819ac4d8a	72	71	72	75	78	75	71	74	77	76	74	72	77	82	80	81	85	86	59	61	45	43	44	45	86	73	80	67
11	6664b0106c4f8ba1febc5a1a	17	30	21	8	14	9	78	68	75	0	73	91	23	30	4	38	22	22	36	73	81	0	91	41	74	14	69	30
84	6665899df8740e0a2b6798fa	16	8	11	6	10	14	89	93	86	88	91	88	15	5	12	10	10	8	94	90	87	90	92	88	91	75	88	79
19	666667f3e900925d27fe40d7	11	11	3	5	5	7	91	75	100	81	84	87	94	94	87	90	100	95	79	88	92	90	88	100	100	98	87	93
26	6669a1aedfa639b507eccff6	79	66	77	85	67	76	26	21	16	36	38	23	84	76	78	88	85	66	60	62	56	69	61	66	24	36	35	25

90 rows × 29 columns

prep for socioecnomic and big5 scores

big5_file_loc = 'raw/big5_scores.csv'
big5_df = pd.read_csv(big5_file_loc)
big5_df.tail(5)

	ResponseId	Extraversion	Agreeableness	Conscientiousness	Neuroticism	Open_Mindedness
99	60a062ed4c4334abbbf32323	3.333333	3.000000	3.166667	3.666667	4.166667
100	5f3ac1732efa0a74f975b1a8	3.500000	4.833333	4.333333	1.000000	3.000000
101	6601b51675d287d6a62f11c4	3.666667	4.833333	4.333333	1.166667	4.666667
102	656f2e99bd2939f9b3f9d090	3.833333	4.000000	4.666667	1.666667	4.000000
103	6658822c0f5b1367a1585ee1	4.500000	4.666667	5.000000	1.333333	3.166667

socioecono_file_loc = 'raw/socioecono_scores.csv'
socioecono_df = pd.read_csv(socioecono_file_loc)
socioecono_df.tail(5)

	ResponseId	gender	age	education	race	employment	income
99	60a062ed4c4334abbbf32323	2	27	4	2	1	1
100	5f3ac1732efa0a74f975b1a8	2	30	2	1	6	5
101	6601b51675d287d6a62f11c4	2	55	5	2	99	11
102	656f2e99bd2939f9b3f9d090	1	43	5	2	7	12
103	6658822c0f5b1367a1585ee1	2	43	5	2	7	8

# filter socioecono_df and big5_df to study3 participants
socioecono_df = socioecono_df[socioecono_df['ResponseId'].isin(target_respondants)]
big5_df = big5_df[big5_df['ResponseId'].isin(target_respondants)]
print(f"row # for socioecono_df : {len(socioecono_df['ResponseId'])}, row # for big5_df : {len(big5_df['ResponseId'])}")

row # for socioecono_df : 90, row # for big5_df : 90

general analysis

print(f"mean age {socioecono_df['age'].mean()} with sd {socioecono_df['age'].std()}, female % {socioecono_df[socioecono_df['gender']==1]['gender'].sum()/socioecono_df.shape[0]}")

mean age 40.86666666666667 with sd 11.74466170869495, female % 0.4777777777777778

# Melt the DataFrame
big5_df_num = big5_df.iloc[:,1:]
big5_df_melted = big5_df_num.melt(var_name='Trait', value_name='Score')

# Create a violin plot
plt.figure(figsize=(10, 6))
sns.violinplot(x='Trait', y='Score', data=big5_df_melted, inner='quartile', palette='muted')
plt.title('Distribution of Scores for Each Trait')
plt.ylim(0, 5)
plt.xlabel('Trait')
plt.ylabel('Score')
plt.grid(True)

# Show the plot
plt.show()

generating distance_df for mixed personality trait

# Extract rows where all Big 5 scores are greater than 3
big5_df_high_scorers = big5_df[
                               (big5_df['Extraversion'] > 3) & 
                               (big5_df['Agreeableness'] > 3) & 
                               (big5_df['Conscientiousness'] > 3) & 
                               (big5_df['Open_Mindedness'] > 3)
]

# Display the result
print(f"Number of participants with targeted Big 5 scores > 3: {len(big5_df_high_scorers)}")
#print(big5_df_high_scorers)

Number of participants with targeted Big 5 scores > 3: 32

# Extract rows where all Big 5 scores are greater than 3
big5_df_ex_high_scorers = big5_df[big5_df['Extraversion'] > 3]['ResponseId'].tolist()
big5_df_ag_high_scorers = big5_df[big5_df['Agreeableness'] > 3]['ResponseId'].tolist()
big5_df_co_high_scorers = big5_df[big5_df['Conscientiousness'] > 3]['ResponseId'].tolist()
big5_df_op_high_scorers = big5_df[big5_df['Open_Mindedness'] > 3]['ResponseId'].tolist()

# Display the result
print(f"Number of participants with op > 3: {len(big5_df_op_high_scorers)}")
print(f"Number of participants with co > 3: {len(big5_df_co_high_scorers)}")
print(f"Number of participants with ex > 3: {len(big5_df_ex_high_scorers)}")
print(f"Number of participants with ag > 3: {len(big5_df_ag_high_scorers)}")

big5_dfs_dict = {}
big5_dfs_dict['op'] = big5_df[big5_df['ResponseId'].isin(big5_df_op_high_scorers)]
big5_dfs_dict['co'] = big5_df[big5_df['ResponseId'].isin(big5_df_co_high_scorers)]
big5_dfs_dict['ex'] = big5_df[big5_df['ResponseId'].isin(big5_df_ex_high_scorers)]
big5_dfs_dict['ag'] = big5_df[big5_df['ResponseId'].isin(big5_df_ag_high_scorers)]

Number of participants with op > 3: 75
Number of participants with co > 3: 77
Number of participants with ex > 3: 46
Number of participants with ag > 3: 79

socioecono_df_high_op = socioecono_df[socioecono_df['ResponseId'].isin(big5_df_op_high_scorers)]
socioecono_df_high_co = socioecono_df[socioecono_df['ResponseId'].isin(big5_df_co_high_scorers)]
socioecono_df_high_ex = socioecono_df[socioecono_df['ResponseId'].isin(big5_df_ex_high_scorers)]
socioecono_df_high_ag = socioecono_df[socioecono_df['ResponseId'].isin(big5_df_ag_high_scorers)]


print(f"OP: mean age {socioecono_df_high_op['age'].mean()} with sd {socioecono_df_high_op['age'].std()}, female % {socioecono_df_high_op[socioecono_df_high_op['gender']==1]['gender'].sum()/socioecono_df_high_op.shape[0]}")
print(f"CO: mean age {socioecono_df_high_co['age'].mean()} with sd {socioecono_df_high_co['age'].std()}, female % {socioecono_df_high_co[socioecono_df_high_co['gender']==1]['gender'].sum()/socioecono_df_high_co.shape[0]}")
print(f"EX: mean age {socioecono_df_high_ex['age'].mean()} with sd {socioecono_df_high_ex['age'].std()}, female % {socioecono_df_high_ex[socioecono_df_high_ex['gender']==1]['gender'].sum()/socioecono_df_high_ex.shape[0]}")
print(f"AG: mean age {socioecono_df_high_ag['age'].mean()} with sd {socioecono_df_high_ag['age'].std()}, female % {socioecono_df_high_ag[socioecono_df_high_ag['gender']==1]['gender'].sum()/socioecono_df_high_ag.shape[0]}")

socioecono_dfs_dict = {}
socioecono_dfs_dict['op'] = socioecono_df_high_op
socioecono_dfs_dict['co'] = socioecono_df_high_co
socioecono_dfs_dict['ex'] = socioecono_df_high_ex
socioecono_dfs_dict['ag'] = socioecono_df_high_ag

OP: mean age 40.653333333333336 with sd 11.715655191946228, female % 0.44
CO: mean age 41.675324675324674 with sd 12.262954008379054, female % 0.4935064935064935
EX: mean age 41.608695652173914 with sd 11.408531429241043, female % 0.5652173913043478
AG: mean age 40.40506329113924 with sd 11.88461512201834, female % 0.46835443037974683

mixed_trait_scores = pd.read_csv('raw/mixed_trait_scores.csv')
mixed_trait_scores

	Extraversion	Agreeableness	Conscientiousness	Open_Mindedness
0	3.166667	4.0	4.0	3.833333

big5_dfs_dict['op']

	ResponseId	Extraversion	Agreeableness	Conscientiousness	Neuroticism	Open_Mindedness
1	641357b26802e397b5627d5e	3.833333	3.166667	4.166667	1.833333	4.166667
2	66294a585f1cf3fdaeb80120	4.166667	3.166667	4.666667	1.833333	3.166667
3	661571d6cc16ee34676734df	3.166667	3.333333	3.000000	4.000000	4.166667
4	6629f95a14b6006e6a4b0aef	3.000000	4.333333	3.500000	2.666667	4.166667
5	66463d8b1984a328f78252d7	3.166667	3.666667	2.333333	3.500000	5.000000
...	...	...	...	...	...	...
94	666667f3e900925d27fe40d7	1.833333	2.000000	2.666667	4.333333	4.666667
95	63c0e1b1ee5ca75f61e3b2e1	4.500000	3.333333	3.833333	1.500000	4.000000
96	5d4a6459061e2a000138bad3	3.666667	4.333333	4.666667	2.166667	5.000000
101	6601b51675d287d6a62f11c4	3.666667	4.833333	4.333333	1.166667	4.666667
103	6658822c0f5b1367a1585ee1	4.500000	4.666667	5.000000	1.333333	3.166667

75 rows × 6 columns

big5_dict = {'op': 'Open_Mindedness',
             'co': 'Conscientiousness',
             'ex': 'Extraversion',
             'ag': 'Agreeableness'}
    

def calculate_distance(df, mixed_scores, trait):
    trait_full = big5_dict[trait]
    return np.abs(df[trait_full] - mixed_scores[trait_full].values[0])

def calculate_distance_from_max(df, trait):
    trait_full = big5_dict[trait]
    return 5 - df[trait_full]

def generate_big5_distance_df(big5_df, mixed_trait_scores):
    distances = pd.DataFrame(index=big5_df.index)
    
    for trait in big5_dict.keys():
        distances[f'{trait}_distance'] = calculate_distance(big5_df, mixed_trait_scores, trait)
        distances[f'{trait}_distance_from_max'] = calculate_distance_from_max(big5_df, trait)
    
    distances['total_distance'] = distances[[f'{trait}_distance' for trait in big5_dict.keys()]].sum(axis=1)
    
    distance_columns = [f'{trait}_distance' for trait in big5_dict.keys()]
    distances['average_distance'] = distances[distance_columns].mean(axis=1)
    
    return pd.concat([big5_df['ResponseId'], distances], axis=1)

def generate_big5_distance_type_df(df, mixed_trait_scores):
    distances = pd.DataFrame(index=df.index)
    
    for trait in big5_dict.keys():
        distances[f'{trait}_distance'] = calculate_distance(df, mixed_trait_scores, trait)
        distances[f'{trait}_distance_from_max'] = calculate_distance_from_max(df, trait)
    
    distances['total_distance'] = distances[[f'{trait}_distance' for trait in big5_dict.keys()]].sum(axis=1)
    
    distance_columns = [f'{trait}_distance' for trait in big5_dict.keys()]
    distances['average_distance'] = distances[distance_columns].mean(axis=1)
    
    return pd.concat([df['ResponseId'], distances], axis=1)

# Generate big5_distance_df for all respondents
big5_distance_df = generate_big5_distance_df(big5_df, mixed_trait_scores)

# Generate big5_distance_{type} for each type
big5_distance_dfs_dict = {}
for type, df in big5_dfs_dict.items():
    big5_distance_dfs_dict[type] = generate_big5_distance_type_df(df, mixed_trait_scores)
    globals()[f'big5_distance_{type}'] = big5_distance_dfs_dict[type]

# Display summary statistics for the comprehensive DataFrame
print("Summary for all responses:")
print(big5_distance_df.describe())
print("\n")

# Display top 5 closest participants based on total distance for all responses
print("Top 5 closest participants for all responses based on total distance:")
print(big5_distance_df.sort_values('total_distance').head())
print("\n" + "="*50 + "\n")

# Display summary statistics and top 5 closest participants for each type
for type, df in big5_distance_dfs_dict.items():
    print(f"Summary for {type.upper()}:")
    print(df.describe())
    print("\n")
    
    print(f"Top 5 closest participants for {type.upper()} based on total distance:")
    print(df.sort_values('total_distance').head())
    print("\n" + "="*50 + "\n")

Summary for all responses:
       op_distance  op_distance_from_max  co_distance  co_distance_from_max  \
count    90.000000             90.000000    90.000000             90.000000   
mean      0.587037              1.235185     0.609259              1.038889   
std       0.440766              0.733493     0.423381              0.743700   
min       0.000000              0.000000     0.000000              0.000000   
25%       0.333333              0.666667     0.333333              0.500000   
50%       0.500000              1.166667     0.500000              1.000000   
75%       0.833333              1.791667     0.833333              1.500000   
max       2.666667              3.833333     2.166667              3.166667   

       ex_distance  ex_distance_from_max  ag_distance  ag_distance_from_max  \
count    90.000000             90.000000    90.000000             90.000000   
mean      0.694444              1.920370     0.585185              1.081481   
std       0.509124              0.859778     0.392059              0.702343   
min       0.000000              0.000000     0.000000              0.000000   
25%       0.333333              1.333333     0.333333              0.500000   
50%       0.500000              1.833333     0.666667              1.000000   
75%       1.000000              2.458333     0.833333              1.666667   
max       1.833333              3.666667     2.000000              3.000000   

       total_distance  average_distance  
count       90.000000         90.000000  
mean         2.475926          0.618981  
std          0.988246          0.247061  
min          0.500000          0.125000  
25%          1.833333          0.458333  
50%          2.333333          0.583333  
75%          3.166667          0.791667  
max          5.500000          1.375000  


Top 5 closest participants for all responses based on total distance:
                  ResponseId  op_distance  op_distance_from_max  co_distance  \
80  6665899df8740e0a2b6798fa     0.000000              1.166667     0.000000   
22  663192b96ab56cf1635615b7     0.000000              1.166667     0.166667   
10  661f1144d35ac5240e53ad3f     0.166667              1.333333     0.000000   
92  65aead8916eb9b7185ed25f3     0.500000              0.666667     0.500000   
71  6664b0106c4f8ba1febc5a1a     0.166667              1.000000     0.333333   

    co_distance_from_max  ex_distance  ex_distance_from_max  ag_distance  \
80              1.000000     0.166667              2.000000     0.333333   
22              0.833333     0.666667              2.500000     0.000000   
10              1.000000     0.500000              1.333333     0.166667   
92              0.500000     0.000000              1.833333     0.000000   
71              0.666667     0.166667              1.666667     0.333333   

    ag_distance_from_max  total_distance  average_distance  
80              0.666667        0.500000          0.125000  
22              1.000000        0.833333          0.208333  
10              0.833333        0.833333          0.208333  
92              1.000000        1.000000          0.250000  
71              1.333333        1.000000          0.250000  

==================================================

Summary for OP:
       op_distance  op_distance_from_max  co_distance  co_distance_from_max  \
count    75.000000             75.000000    75.000000             75.000000   
mean      0.466667              1.011111     0.600000              1.000000   
std       0.305112              0.537856     0.427121              0.739795   
min       0.000000              0.000000     0.000000              0.000000   
25%       0.166667              0.666667     0.333333              0.416667   
50%       0.500000              1.000000     0.500000              1.000000   
75%       0.666667              1.500000     0.833333              1.500000   
max       1.166667              1.833333     2.166667              3.166667   

       ex_distance  ex_distance_from_max  ag_distance  ag_distance_from_max  \
count    75.000000             75.000000    75.000000             75.000000   
mean      0.682222              1.826667     0.582222              1.102222   
std       0.494848              0.846491     0.400238              0.702256   
min       0.000000              0.000000     0.000000              0.000000   
25%       0.333333              1.166667     0.333333              0.583333   
50%       0.500000              1.833333     0.666667              1.000000   
75%       1.000000              2.333333     0.833333              1.666667   
max       1.833333              3.666667     2.000000              3.000000   

       total_distance  average_distance  
count       75.000000         75.000000  
mean         2.331111          0.582778  
std          0.944596          0.236149  
min          0.500000          0.125000  
25%          1.750000          0.437500  
50%          2.166667          0.541667  
75%          3.000000          0.750000  
max          5.500000          1.375000  


Top 5 closest participants for OP based on total distance:
                  ResponseId  op_distance  op_distance_from_max  co_distance  \
80  6665899df8740e0a2b6798fa     0.000000              1.166667     0.000000   
22  663192b96ab56cf1635615b7     0.000000              1.166667     0.166667   
10  661f1144d35ac5240e53ad3f     0.166667              1.333333     0.000000   
92  65aead8916eb9b7185ed25f3     0.500000              0.666667     0.500000   
71  6664b0106c4f8ba1febc5a1a     0.166667              1.000000     0.333333   

    co_distance_from_max  ex_distance  ex_distance_from_max  ag_distance  \
80              1.000000     0.166667              2.000000     0.333333   
22              0.833333     0.666667              2.500000     0.000000   
10              1.000000     0.500000              1.333333     0.166667   
92              0.500000     0.000000              1.833333     0.000000   
71              0.666667     0.166667              1.666667     0.333333   

    ag_distance_from_max  total_distance  average_distance  
80              0.666667        0.500000          0.125000  
22              1.000000        0.833333          0.208333  
10              0.833333        0.833333          0.208333  
92              1.000000        1.000000          0.250000  
71              1.333333        1.000000          0.250000  

==================================================

Summary for CO:
       op_distance  op_distance_from_max  co_distance  co_distance_from_max  \
count    77.000000             77.000000    77.000000             77.000000   
mean      0.577922              1.233766     0.493506              0.826840   
std       0.451269              0.733124     0.311149              0.559632   
min       0.000000              0.000000     0.000000              0.000000   
25%       0.333333              0.666667     0.333333              0.333333   
50%       0.500000              1.166667     0.500000              0.833333   
75%       0.833333              1.666667     0.833333              1.333333   
max       2.666667              3.833333     1.000000              1.833333   

       ex_distance  ex_distance_from_max  ag_distance  ag_distance_from_max  \
count    77.000000             77.000000    77.000000             77.000000   
mean      0.696970              1.867965     0.571429              0.982684   
std       0.495393              0.858113     0.352962              0.674615   
min       0.000000              0.000000     0.000000              0.000000   
25%       0.333333              1.333333     0.333333              0.333333   
50%       0.500000              1.833333     0.666667              0.833333   
75%       1.000000              2.333333     0.833333              1.500000   
max       1.833333              3.666667     1.833333              2.833333   

       total_distance  average_distance  
count       77.000000         77.000000  
mean         2.339827          0.584957  
std          0.915446          0.228862  
min          0.500000          0.125000  
25%          1.666667          0.416667  
50%          2.166667          0.541667  
75%          3.000000          0.750000  
max          4.333333          1.083333  


Top 5 closest participants for CO based on total distance:
                  ResponseId  op_distance  op_distance_from_max  co_distance  \
80  6665899df8740e0a2b6798fa     0.000000              1.166667     0.000000   
22  663192b96ab56cf1635615b7     0.000000              1.166667     0.166667   
10  661f1144d35ac5240e53ad3f     0.166667              1.333333     0.000000   
71  6664b0106c4f8ba1febc5a1a     0.166667              1.000000     0.333333   
92  65aead8916eb9b7185ed25f3     0.500000              0.666667     0.500000   

    co_distance_from_max  ex_distance  ex_distance_from_max  ag_distance  \
80              1.000000     0.166667              2.000000     0.333333   
22              0.833333     0.666667              2.500000     0.000000   
10              1.000000     0.500000              1.333333     0.166667   
71              0.666667     0.166667              1.666667     0.333333   
92              0.500000     0.000000              1.833333     0.000000   

    ag_distance_from_max  total_distance  average_distance  
80              0.666667        0.500000          0.125000  
22              1.000000        0.833333          0.208333  
10              0.833333        0.833333          0.208333  
71              1.333333        1.000000          0.250000  
92              1.000000        1.000000          0.250000  

==================================================

Summary for EX:
       op_distance  op_distance_from_max  co_distance  co_distance_from_max  \
count    46.000000             46.000000    46.000000             46.000000   
mean      0.547101              1.090580     0.550725              0.833333   
std       0.328977              0.638968     0.383083              0.654519   
min       0.000000              0.000000     0.000000              0.000000   
25%       0.333333              0.666667     0.333333              0.333333   
50%       0.500000              1.000000     0.500000              0.666667   
75%       0.833333              1.666667     0.833333              1.166667   
max       1.166667              2.333333     1.666667              2.666667   

       ex_distance  ex_distance_from_max  ag_distance  ag_distance_from_max  \
count    46.000000             46.000000    46.000000             46.000000   
mean      0.594203              1.239130     0.634058              1.039855   
std       0.473563              0.473563     0.377810              0.743021   
min       0.000000              0.000000     0.000000              0.000000   
25%       0.166667              0.833333     0.333333              0.333333   
50%       0.500000              1.333333     0.666667              1.000000   
75%       1.000000              1.666667     0.833333              1.666667   
max       1.833333              1.833333     1.833333              2.833333   

       total_distance  average_distance  
count       46.000000         46.000000  
mean         2.326087          0.581522  
std          0.859914          0.214979  
min          0.833333          0.208333  
25%          1.833333          0.458333  
50%          2.333333          0.583333  
75%          2.916667          0.729167  
max          4.333333          1.083333  


Top 5 closest participants for EX based on total distance:
                  ResponseId  op_distance  op_distance_from_max  co_distance  \
10  661f1144d35ac5240e53ad3f     0.166667              1.333333     0.000000   
92  65aead8916eb9b7185ed25f3     0.500000              0.666667     0.500000   
71  6664b0106c4f8ba1febc5a1a     0.166667              1.000000     0.333333   
72  5c51d7f198b0ff000110dba8     0.000000              1.166667     0.333333   
6   5d215a1bbf7f840019701939     0.833333              0.333333     0.166667   

    co_distance_from_max  ex_distance  ex_distance_from_max  ag_distance  \
10              1.000000     0.500000              1.333333     0.166667   
92              0.500000     0.000000              1.833333     0.000000   
71              0.666667     0.166667              1.666667     0.333333   
72              0.666667     0.333333              1.500000     0.333333   
6               0.833333     0.000000              1.833333     0.000000   

    ag_distance_from_max  total_distance  average_distance  
10              0.833333        0.833333          0.208333  
92              1.000000        1.000000          0.250000  
71              1.333333        1.000000          0.250000  
72              1.333333        1.000000          0.250000  
6               1.000000        1.000000          0.250000  

==================================================

Summary for AG:
       op_distance  op_distance_from_max  co_distance  co_distance_from_max  \
count    79.000000             79.000000    79.000000             79.000000   
mean      0.597046              1.236287     0.609705              1.010549   
std       0.457299              0.751829     0.409021              0.737354   
min       0.000000              0.000000     0.000000              0.000000   
25%       0.333333              0.666667     0.333333              0.333333   
50%       0.500000              1.166667     0.500000              1.000000   
75%       0.833333              1.833333     0.833333              1.500000   
max       2.666667              3.833333     2.166667              3.166667   

       ex_distance  ex_distance_from_max  ag_distance  ag_distance_from_max  \
count    79.000000             79.000000    79.000000             79.000000   
mean      0.704641              1.930380     0.491561              0.917722   
std       0.524302              0.876492     0.294657              0.569821   
min       0.000000              0.000000     0.000000              0.000000   
25%       0.250000              1.333333     0.333333              0.333333   
50%       0.666667              1.833333     0.500000              1.000000   
75%       1.000000              2.500000     0.666667              1.416667   
max       1.833333              3.666667     1.000000              1.833333   

       total_distance  average_distance  
count       79.000000         79.000000  
mean         2.402954          0.600738  
std          0.943416          0.235854  
min          0.500000          0.125000  
25%          1.833333          0.458333  
50%          2.333333          0.583333  
75%          3.166667          0.791667  
max          4.333333          1.083333  


Top 5 closest participants for AG based on total distance:
                  ResponseId  op_distance  op_distance_from_max  co_distance  \
80  6665899df8740e0a2b6798fa     0.000000              1.166667     0.000000   
22  663192b96ab56cf1635615b7     0.000000              1.166667     0.166667   
10  661f1144d35ac5240e53ad3f     0.166667              1.333333     0.000000   
92  65aead8916eb9b7185ed25f3     0.500000              0.666667     0.500000   
71  6664b0106c4f8ba1febc5a1a     0.166667              1.000000     0.333333   

    co_distance_from_max  ex_distance  ex_distance_from_max  ag_distance  \
80              1.000000     0.166667              2.000000     0.333333   
22              0.833333     0.666667              2.500000     0.000000   
10              1.000000     0.500000              1.333333     0.166667   
92              0.500000     0.000000              1.833333     0.000000   
71              0.666667     0.166667              1.666667     0.333333   

    ag_distance_from_max  total_distance  average_distance  
80              0.666667        0.500000          0.125000  
22              1.000000        0.833333          0.208333  
10              0.833333        0.833333          0.208333  
92              1.000000        1.000000          0.250000  
71              1.333333        1.000000          0.250000  

==================================================

       op_distance  op_distance_from_max  co_distance  co_distance_from_max  \
count    79.000000             79.000000    79.000000             79.000000   
mean      0.597046              1.236287     0.609705              1.010549   
std       0.457299              0.751829     0.409021              0.737354   
min       0.000000              0.000000     0.000000              0.000000   
25%       0.333333              0.666667     0.333333              0.333333   
50%       0.500000              1.166667     0.500000              1.000000   
75%       0.833333              1.833333     0.833333              1.500000   
max       2.666667              3.833333     2.166667              3.166667   

       ex_distance  ex_distance_from_max  ag_distance  ag_distance_from_max  \
count    79.000000             79.000000    79.000000             79.000000   
mean      0.704641              1.930380     0.491561              0.917722   
std       0.524302              0.876492     0.294657              0.569821   
min       0.000000              0.000000     0.000000              0.000000   
25%       0.250000              1.333333     0.333333              0.333333   
50%       0.666667              1.833333     0.500000              1.000000   
75%       1.000000              2.500000     0.666667              1.416667   
max       1.833333              3.666667     1.000000              1.833333   

       total_distance  average_distance  
count       79.000000         79.000000  
mean         2.402954          0.600738  
std          0.943416          0.235854  
min          0.500000          0.125000  
25%          1.833333          0.458333  
50%          2.333333          0.583333  
75%          3.166667          0.791667  
max          4.333333          1.083333  


Top 5 closest participants for AG based on total distance:
                  ResponseId  op_distance  op_distance_from_max  co_distance  \
80  6665899df8740e0a2b6798fa     0.000000              1.166667     0.000000   
22  663192b96ab56cf1635615b7     0.000000              1.166667     0.166667   
10  661f1144d35ac5240e53ad3f     0.166667              1.333333     0.000000   
92  65aead8916eb9b7185ed25f3     0.500000              0.666667     0.500000   
71  6664b0106c4f8ba1febc5a1a     0.166667              1.000000     0.333333   

    co_distance_from_max  ex_distance  ex_distance_from_max  ag_distance  \
80              1.000000     0.166667              2.000000     0.333333   
22              0.833333     0.666667              2.500000     0.000000   
10              1.000000     0.500000              1.333333     0.166667   
92              0.500000     0.000000              1.833333     0.000000   
71              0.666667     0.166667              1.666667     0.333333   

    ag_distance_from_max  total_distance  average_distance  
80              0.666667        0.500000          0.125000  
22              1.000000        0.833333          0.208333  
10              0.833333        0.833333          0.208333  
92              1.000000        1.000000          0.250000  
71              1.333333        1.000000          0.250000  

==================================================

for key, value in big5_distance_dfs_dict.items():
    print(key)
    print(value.head())
    print("")

op
                 ResponseId  op_distance  op_distance_from_max  co_distance  \
1  641357b26802e397b5627d5e     0.333333              0.833333     0.166667   
2  66294a585f1cf3fdaeb80120     0.666667              1.833333     0.666667   
3  661571d6cc16ee34676734df     0.333333              0.833333     1.000000   
4  6629f95a14b6006e6a4b0aef     0.333333              0.833333     0.500000   
5  66463d8b1984a328f78252d7     1.166667              0.000000     1.666667   

   co_distance_from_max  ex_distance  ex_distance_from_max  ag_distance  \
1              0.833333     0.666667              1.166667     0.833333   
2              0.333333     1.000000              0.833333     0.833333   
3              2.000000     0.000000              1.833333     0.666667   
4              1.500000     0.166667              2.000000     0.333333   
5              2.666667     0.000000              1.833333     0.333333   

   ag_distance_from_max  total_distance  average_distance  
1              1.833333        2.000000          0.500000  
2              1.833333        3.166667          0.791667  
3              1.666667        2.000000          0.500000  
4              0.666667        1.333333          0.333333  
5              1.333333        3.166667          0.791667  

co
                 ResponseId  op_distance  op_distance_from_max  co_distance  \
1  641357b26802e397b5627d5e     0.333333              0.833333     0.166667   
2  66294a585f1cf3fdaeb80120     0.666667              1.833333     0.666667   
4  6629f95a14b6006e6a4b0aef     0.333333              0.833333     0.500000   
6  5d215a1bbf7f840019701939     0.833333              0.333333     0.166667   
8  6601c47dbc5b1b7c9f1b3fde     0.333333              1.500000     0.166667   

   co_distance_from_max  ex_distance  ex_distance_from_max  ag_distance  \
1              0.833333     0.666667              1.166667     0.833333   
2              0.333333     1.000000              0.833333     0.833333   
4              1.500000     0.166667              2.000000     0.333333   
6              0.833333     0.000000              1.833333     0.000000   
8              0.833333     0.500000              2.333333     0.000000   

   ag_distance_from_max  total_distance  average_distance  
1              1.833333        2.000000          0.500000  
2              1.833333        3.166667          0.791667  
4              0.666667        1.333333          0.333333  
6              1.000000        1.000000          0.250000  
8              1.000000        1.000000          0.250000  

ex
                 ResponseId  op_distance  op_distance_from_max  co_distance  \
1  641357b26802e397b5627d5e     0.333333              0.833333     0.166667   
2  66294a585f1cf3fdaeb80120     0.666667              1.833333     0.666667   
3  661571d6cc16ee34676734df     0.333333              0.833333     1.000000   
5  66463d8b1984a328f78252d7     1.166667              0.000000     1.666667   
6  5d215a1bbf7f840019701939     0.833333              0.333333     0.166667   

   co_distance_from_max  ex_distance  ex_distance_from_max  ag_distance  \
1              0.833333     0.666667              1.166667     0.833333   
2              0.333333     1.000000              0.833333     0.833333   
3              2.000000     0.000000              1.833333     0.666667   
5              2.666667     0.000000              1.833333     0.333333   
6              0.833333     0.000000              1.833333     0.000000   

   ag_distance_from_max  total_distance  average_distance  
1              1.833333        2.000000          0.500000  
2              1.833333        3.166667          0.791667  
3              1.666667        2.000000          0.500000  
5              1.333333        3.166667          0.791667  
6              1.000000        1.000000          0.250000  

ag
                 ResponseId  op_distance  op_distance_from_max  co_distance  \
1  641357b26802e397b5627d5e     0.333333              0.833333     0.166667   
2  66294a585f1cf3fdaeb80120     0.666667              1.833333     0.666667   
3  661571d6cc16ee34676734df     0.333333              0.833333     1.000000   
4  6629f95a14b6006e6a4b0aef     0.333333              0.833333     0.500000   
5  66463d8b1984a328f78252d7     1.166667              0.000000     1.666667   

   co_distance_from_max  ex_distance  ex_distance_from_max  ag_distance  \
1              0.833333     0.666667              1.166667     0.833333   
2              0.333333     1.000000              0.833333     0.833333   
3              2.000000     0.000000              1.833333     0.666667   
4              1.500000     0.166667              2.000000     0.333333   
5              2.666667     0.000000              1.833333     0.333333   

   ag_distance_from_max  total_distance  average_distance  
1              1.833333        2.000000          0.500000  
2              1.833333        3.166667          0.791667  
3              1.666667        2.000000          0.500000  
4              0.666667        1.333333          0.333333  
5              1.333333        3.166667          0.791667  

from_max_cols = ['op_distance_from_max',
                 'co_distance_from_max',
                 'ex_distance_from_max',
                 'ag_distance_from_max']

big5_distance_dfs_dict['op'][from_max_cols[0]].hist()

#big5_distance_dfs_dict['op']

<Axes: >

op_filtered_rows = big5_distance_dfs_dict['op'][big5_distance_dfs_dict['op'][from_max_cols[0]] < 1]
co_filtered_rows = big5_distance_dfs_dict['co'][big5_distance_dfs_dict['co'][from_max_cols[1]] < 1]
ex_filtered_rows = big5_distance_dfs_dict['ex'][big5_distance_dfs_dict['ex'][from_max_cols[2]] < 1]
ag_filtered_rows = big5_distance_dfs_dict['ag'][big5_distance_dfs_dict['ag'][from_max_cols[3]] < 1]

print(f"op: {op_filtered_rows.shape[0]}")
print(f"co: {co_filtered_rows.shape[0]}")
print(f"ex: {ex_filtered_rows.shape[0]}")
print(f"ag: {ag_filtered_rows.shape[0]}")

op: 35
co: 42
ex: 15
ag: 39

co: 42
ex: 15
ag: 39

big5_distance_df['average_distance'].hist()

<Axes: >

big5_distance_dfs_dict['op']

	ResponseId	op_distance	op_distance_from_max	co_distance	co_distance_from_max	ex_distance	ex_distance_from_max	ag_distance	ag_distance_from_max	total_distance	average_distance
1	641357b26802e397b5627d5e	0.333333	0.833333	0.166667	0.833333	0.666667	1.166667	0.833333	1.833333	2.000000	0.500000
2	66294a585f1cf3fdaeb80120	0.666667	1.833333	0.666667	0.333333	1.000000	0.833333	0.833333	1.833333	3.166667	0.791667
3	661571d6cc16ee34676734df	0.333333	0.833333	1.000000	2.000000	0.000000	1.833333	0.666667	1.666667	2.000000	0.500000
4	6629f95a14b6006e6a4b0aef	0.333333	0.833333	0.500000	1.500000	0.166667	2.000000	0.333333	0.666667	1.333333	0.333333
5	66463d8b1984a328f78252d7	1.166667	0.000000	1.666667	2.666667	0.000000	1.833333	0.333333	1.333333	3.166667	0.791667
...	...	...	...	...	...	...	...	...	...	...	...
94	666667f3e900925d27fe40d7	0.833333	0.333333	1.333333	2.333333	1.333333	3.166667	2.000000	3.000000	5.500000	1.375000
95	63c0e1b1ee5ca75f61e3b2e1	0.166667	1.000000	0.166667	1.166667	1.333333	0.500000	0.666667	1.666667	2.333333	0.583333
96	5d4a6459061e2a000138bad3	1.166667	0.000000	0.666667	0.333333	0.500000	1.333333	0.333333	0.666667	2.666667	0.666667
101	6601b51675d287d6a62f11c4	0.833333	0.333333	0.333333	0.666667	0.500000	1.333333	0.833333	0.166667	2.500000	0.625000
103	6658822c0f5b1367a1585ee1	0.666667	1.833333	1.000000	0.000000	1.333333	0.500000	0.666667	0.333333	3.666667	0.916667

75 rows × 11 columns

data prep for mixed effect model

big5_distance_df

	ResponseId	op_distance	op_distance_from_max	co_distance	co_distance_from_max	ex_distance	ex_distance_from_max	ag_distance	ag_distance_from_max	total_distance	average_distance
1	641357b26802e397b5627d5e	0.333333	0.833333	0.166667	0.833333	0.666667	1.166667	0.833333	1.833333	2.000000	0.500000
2	66294a585f1cf3fdaeb80120	0.666667	1.833333	0.666667	0.333333	1.000000	0.833333	0.833333	1.833333	3.166667	0.791667
3	661571d6cc16ee34676734df	0.333333	0.833333	1.000000	2.000000	0.000000	1.833333	0.666667	1.666667	2.000000	0.500000
4	6629f95a14b6006e6a4b0aef	0.333333	0.833333	0.500000	1.500000	0.166667	2.000000	0.333333	0.666667	1.333333	0.333333
5	66463d8b1984a328f78252d7	1.166667	0.000000	1.666667	2.666667	0.000000	1.833333	0.333333	1.333333	3.166667	0.791667
...	...	...	...	...	...	...	...	...	...	...	...
96	5d4a6459061e2a000138bad3	1.166667	0.000000	0.666667	0.333333	0.500000	1.333333	0.333333	0.666667	2.666667	0.666667
98	62b473736986d2b27c1427d2	1.166667	2.333333	0.333333	1.333333	1.000000	0.833333	0.500000	1.500000	3.000000	0.750000
100	5f3ac1732efa0a74f975b1a8	0.833333	2.000000	0.333333	0.666667	0.333333	1.500000	0.833333	0.166667	2.333333	0.583333
101	6601b51675d287d6a62f11c4	0.833333	0.333333	0.333333	0.666667	0.500000	1.333333	0.833333	0.166667	2.500000	0.625000
103	6658822c0f5b1367a1585ee1	0.666667	1.833333	1.000000	0.000000	1.333333	0.500000	0.666667	0.333333	3.666667	0.916667

90 rows × 11 columns

s3_response_df

	ResponseId	Ads_OP_1	Ads_OP_2	Ads_OP_3	Ads_OP_4	Ads_OP_5	Ads_OP_6	Ads_CO_1	Ads_CO_2	Ads_CO_3	Ads_CO_4	Ads_CO_5	Ads_CO_6	Ads_EX_1	Ads_EX_2	Ads_EX_3	Ads_EX_4	Ads_EX_5	Ads_EX_6	Ads_AG_1	Ads_AG_2	Ads_AG_3	Ads_AG_4	Ads_AG_5	Ads_AG_6	Desc_OP_1	Desc_CO_1	Desc_EX_1	Desc_AG_1
12	577f70d1454e5400015f1845	72	59	70	34	88	83	88	85	91	87	95	93	7	29	36	5	20	10	82	89	91	88	91	91	91	85	88	6
81	5a2adf6a8e00a000019864fb	72	55	64	67	58	70	27	39	35	35	22	29	74	70	82	77	75	80	80	76	76	66	77	67	72	25	26	71
33	5aa806e7777df200016088c5	20	9	0	0	0	0	16	0	6	0	11	0	16	0	2	10	4	0	62	69	54	38	48	67	84	74	23	59
63	5afe053059ae1e00017e6a3c	78	19	17	16	17	19	0	20	18	62	9	10	47	46	47	27	46	47	44	44	47	47	47	47	83	72	70	64
10	5b757a03976ecb0001b1168f	8	19	0	0	0	0	64	73	73	65	68	70	27	39	14	8	17	7	71	84	77	68	92	93	45	27	81	42
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
82	66645cff5b7cb46819ac4d8a	72	71	72	75	78	75	71	74	77	76	74	72	77	82	80	81	85	86	59	61	45	43	44	45	86	73	80	67
11	6664b0106c4f8ba1febc5a1a	17	30	21	8	14	9	78	68	75	0	73	91	23	30	4	38	22	22	36	73	81	0	91	41	74	14	69	30
84	6665899df8740e0a2b6798fa	16	8	11	6	10	14	89	93	86	88	91	88	15	5	12	10	10	8	94	90	87	90	92	88	91	75	88	79
19	666667f3e900925d27fe40d7	11	11	3	5	5	7	91	75	100	81	84	87	94	94	87	90	100	95	79	88	92	90	88	100	100	98	87	93
26	6669a1aedfa639b507eccff6	79	66	77	85	67	76	26	21	16	36	38	23	84	76	78	88	85	66	60	62	56	69	61	66	24	36	35	25

90 rows × 29 columns

# Merge big5_distance_df and s3_response_df on ResponseId
merged_df = pd.merge(big5_distance_df, s3_response_df, on='ResponseId', how='inner')

# Rename columns in the merged dataframe
column_mapping = {
    'op_distance': 'OP_distance',
    'op_distance_from_max': 'OP_distance_from_max',
    'co_distance': 'CO_distance',
    'co_distance_from_max': 'CO_distance_from_max',
    'ex_distance': 'EX_distance',
    'ex_distance_from_max': 'EX_distance_from_max',
    'ag_distance': 'AG_distance',
    'ag_distance_from_max': 'AG_distance_from_max'
}

merged_df = merged_df.rename(columns=column_mapping)

# Display the column names to verify the changes
print("\nColumn names:")
print(merged_df.columns)

Column names:
Index(['ResponseId', 'OP_distance', 'OP_distance_from_max', 'CO_distance',
       'CO_distance_from_max', 'EX_distance', 'EX_distance_from_max',
       'AG_distance', 'AG_distance_from_max', 'total_distance',
       'average_distance', 'Ads_OP_1', 'Ads_OP_2', 'Ads_OP_3', 'Ads_OP_4',
       'Ads_OP_5', 'Ads_OP_6', 'Ads_CO_1', 'Ads_CO_2', 'Ads_CO_3', 'Ads_CO_4',
       'Ads_CO_5', 'Ads_CO_6', 'Ads_EX_1', 'Ads_EX_2', 'Ads_EX_3', 'Ads_EX_4',
       'Ads_EX_5', 'Ads_EX_6', 'Ads_AG_1', 'Ads_AG_2', 'Ads_AG_3', 'Ads_AG_4',
       'Ads_AG_5', 'Ads_AG_6', 'Desc_OP_1', 'Desc_CO_1', 'Desc_EX_1',
       'Desc_AG_1'],
      dtype='object')

merged_df

	ResponseId	OP_distance	OP_distance_from_max	CO_distance	CO_distance_from_max	EX_distance	EX_distance_from_max	AG_distance	AG_distance_from_max	total_distance	average_distance	Ads_OP_1	Ads_OP_2	Ads_OP_3	Ads_OP_4	Ads_OP_5	Ads_OP_6	Ads_CO_1	Ads_CO_2	Ads_CO_3	Ads_CO_4	Ads_CO_5	Ads_CO_6	Ads_EX_1	Ads_EX_2	Ads_EX_3	Ads_EX_4	Ads_EX_5	Ads_EX_6	Ads_AG_1	Ads_AG_2	Ads_AG_3	Ads_AG_4	Ads_AG_5	Ads_AG_6	Desc_OP_1	Desc_CO_1	Desc_EX_1	Desc_AG_1
0	641357b26802e397b5627d5e	0.333333	0.833333	0.166667	0.833333	0.666667	1.166667	0.833333	1.833333	2.000000	0.500000	89	75	89	84	71	75	81	66	74	77	79	82	72	68	67	68	76	69	71	78	70	68	85	84	63	50	53	50
1	66294a585f1cf3fdaeb80120	0.666667	1.833333	0.666667	0.333333	1.000000	0.833333	0.833333	1.833333	3.166667	0.791667	72	50	53	28	50	59	34	50	50	59	46	21	54	50	42	38	45	62	56	50	55	64	58	60	91	10	80	60
2	661571d6cc16ee34676734df	0.333333	0.833333	1.000000	2.000000	0.000000	1.833333	0.666667	1.666667	2.000000	0.500000	29	49	37	45	34	31	60	60	58	59	62	80	61	100	56	57	61	67	65	59	59	59	65	71	28	49	57	57
3	6629f95a14b6006e6a4b0aef	0.333333	0.833333	0.500000	1.500000	0.166667	2.000000	0.333333	0.666667	1.333333	0.333333	37	37	40	42	40	41	64	66	71	60	66	66	33	34	38	35	36	35	64	62	59	60	62	66	68	62	65	36
4	66463d8b1984a328f78252d7	1.166667	0.000000	1.666667	2.666667	0.000000	1.833333	0.333333	1.333333	3.166667	0.791667	13	39	33	23	40	15	81	73	77	85	61	84	16	21	20	30	14	26	20	40	8	28	18	16	14	94	65	9
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
85	5d4a6459061e2a000138bad3	1.166667	0.000000	0.666667	0.333333	0.500000	1.333333	0.333333	0.666667	2.666667	0.666667	48	73	48	48	48	47	11	40	10	10	9	10	93	52	96	95	95	95	14	80	14	46	17	44	7	26	54	47
86	62b473736986d2b27c1427d2	1.166667	2.333333	0.333333	1.333333	1.000000	0.833333	0.500000	1.500000	3.000000	0.750000	72	27	27	40	17	30	71	88	26	26	68	82	33	15	76	30	70	13	74	26	66	25	68	45	21	28	84	68
87	5f3ac1732efa0a74f975b1a8	0.833333	2.000000	0.333333	0.666667	0.333333	1.500000	0.833333	0.166667	2.333333	0.583333	5	14	0	3	31	0	24	0	26	0	46	0	29	1	3	50	28	23	81	58	77	60	51	59	100	59	0	11
88	6601b51675d287d6a62f11c4	0.833333	0.333333	0.333333	0.666667	0.500000	1.333333	0.833333	0.166667	2.500000	0.625000	18	44	19	23	20	24	33	45	34	40	39	35	18	40	17	17	20	22	71	83	79	75	82	89	45	44	83	57
89	6658822c0f5b1367a1585ee1	0.666667	1.833333	1.000000	0.000000	1.333333	0.500000	0.666667	0.333333	3.666667	0.916667	40	50	40	40	40	40	80	80	80	85	90	90	90	50	90	80	100	90	60	70	60	50	70	70	40	70	60	70

90 rows × 39 columns

merged_df

	ResponseId	OP_distance	OP_distance_from_max	CO_distance	CO_distance_from_max	EX_distance	EX_distance_from_max	AG_distance	AG_distance_from_max	total_distance	average_distance	Ads_OP_1	Ads_OP_2	Ads_OP_3	Ads_OP_4	Ads_OP_5	Ads_OP_6	Ads_CO_1	Ads_CO_2	Ads_CO_3	Ads_CO_4	Ads_CO_5	Ads_CO_6	Ads_EX_1	Ads_EX_2	Ads_EX_3	Ads_EX_4	Ads_EX_5	Ads_EX_6	Ads_AG_1	Ads_AG_2	Ads_AG_3	Ads_AG_4	Ads_AG_5	Ads_AG_6	Desc_OP_1	Desc_CO_1	Desc_EX_1	Desc_AG_1
0	641357b26802e397b5627d5e	0.333333	0.833333	0.166667	0.833333	0.666667	1.166667	0.833333	1.833333	2.000000	0.500000	89	75	89	84	71	75	81	66	74	77	79	82	72	68	67	68	76	69	71	78	70	68	85	84	63	50	53	50
1	66294a585f1cf3fdaeb80120	0.666667	1.833333	0.666667	0.333333	1.000000	0.833333	0.833333	1.833333	3.166667	0.791667	72	50	53	28	50	59	34	50	50	59	46	21	54	50	42	38	45	62	56	50	55	64	58	60	91	10	80	60
2	661571d6cc16ee34676734df	0.333333	0.833333	1.000000	2.000000	0.000000	1.833333	0.666667	1.666667	2.000000	0.500000	29	49	37	45	34	31	60	60	58	59	62	80	61	100	56	57	61	67	65	59	59	59	65	71	28	49	57	57
3	6629f95a14b6006e6a4b0aef	0.333333	0.833333	0.500000	1.500000	0.166667	2.000000	0.333333	0.666667	1.333333	0.333333	37	37	40	42	40	41	64	66	71	60	66	66	33	34	38	35	36	35	64	62	59	60	62	66	68	62	65	36
4	66463d8b1984a328f78252d7	1.166667	0.000000	1.666667	2.666667	0.000000	1.833333	0.333333	1.333333	3.166667	0.791667	13	39	33	23	40	15	81	73	77	85	61	84	16	21	20	30	14	26	20	40	8	28	18	16	14	94	65	9
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
85	5d4a6459061e2a000138bad3	1.166667	0.000000	0.666667	0.333333	0.500000	1.333333	0.333333	0.666667	2.666667	0.666667	48	73	48	48	48	47	11	40	10	10	9	10	93	52	96	95	95	95	14	80	14	46	17	44	7	26	54	47
86	62b473736986d2b27c1427d2	1.166667	2.333333	0.333333	1.333333	1.000000	0.833333	0.500000	1.500000	3.000000	0.750000	72	27	27	40	17	30	71	88	26	26	68	82	33	15	76	30	70	13	74	26	66	25	68	45	21	28	84	68
87	5f3ac1732efa0a74f975b1a8	0.833333	2.000000	0.333333	0.666667	0.333333	1.500000	0.833333	0.166667	2.333333	0.583333	5	14	0	3	31	0	24	0	26	0	46	0	29	1	3	50	28	23	81	58	77	60	51	59	100	59	0	11
88	6601b51675d287d6a62f11c4	0.833333	0.333333	0.333333	0.666667	0.500000	1.333333	0.833333	0.166667	2.500000	0.625000	18	44	19	23	20	24	33	45	34	40	39	35	18	40	17	17	20	22	71	83	79	75	82	89	45	44	83	57
89	6658822c0f5b1367a1585ee1	0.666667	1.833333	1.000000	0.000000	1.333333	0.500000	0.666667	0.333333	3.666667	0.916667	40	50	40	40	40	40	80	80	80	85	90	90	90	50	90	80	100	90	60	70	60	50	70	70	40	70	60	70

90 rows × 39 columns

# Define the columns to normalize
columns_to_normalize = [
    'OP_distance', 'OP_distance_from_max', 'CO_distance', 'CO_distance_from_max',
    'EX_distance', 'EX_distance_from_max', 'AG_distance', 'AG_distance_from_max',
    'total_distance', 'average_distance'
]

# Create a new DataFrame with normalized columns
merged_df_norm = merged_df.copy()

# Normalize the specified columns
for column in columns_to_normalize:
    merged_df_norm[column] = (merged_df_norm[column] - merged_df_norm[column].min()) / (merged_df_norm[column].max() - merged_df_norm[column].min())

merged_df_norm[columns_to_normalize].hist(figsize= (12,10))

array([[<Axes: title={'center': 'OP_distance'}>,
        <Axes: title={'center': 'OP_distance_from_max'}>,
        <Axes: title={'center': 'CO_distance'}>],
       [<Axes: title={'center': 'CO_distance_from_max'}>,
        <Axes: title={'center': 'EX_distance'}>,
        <Axes: title={'center': 'EX_distance_from_max'}>],
       [<Axes: title={'center': 'AG_distance'}>,
        <Axes: title={'center': 'AG_distance_from_max'}>,
        <Axes: title={'center': 'total_distance'}>],
       [<Axes: title={'center': 'average_distance'}>, <Axes: >, <Axes: >]],
      dtype=object)

# Extract rows where both 'EX_distance_from_max' and 'average_distance' are <= 0.7
filtered_df = merged_df_norm[(merged_df_norm['EX_distance_from_max'] <= 0.7) & (merged_df_norm['average_distance'] <= 0.6)]

# Optional: Calculate the percentage of rows meeting the condition
percentage = (len(filtered_df) / len(merged_df_norm)) * 100
print(f"Percentage of rows meeting the condition: {percentage:.2f}%")

Percentage of rows meeting the condition: 70.00%

import pandas as pd
import numpy as np
import statsmodels.api as sm
import statsmodels.formula.api as smf
from scipy import stats
from sklearn.preprocessing import StandardScaler
from statsmodels.stats.outliers_influence import variance_inflation_factor
import matplotlib.pyplot as plt

data = merged_df

# ResponseIdを文字列に変換
data['ResponseId'] = data['ResponseId'].astype(str)

# データの前処理
traits = ['EX', 'OP', 'CO', 'AG']
for trait in traits:
    data[f'{trait}_distance_from_max_scaled'] = StandardScaler().fit_transform(data[[f'{trait}_distance_from_max']])

data['average_distance_scaled'] = StandardScaler().fit_transform(data[['average_distance']])

# 混合効果モデルの構築
def run_mixed_model(data, trait):
    ad_preference = f'Ads_{trait}_1'
    formula = (f"{ad_preference} ~ {trait}_distance_from_max_scaled + average_distance_scaled + "
               f"{trait}_distance_from_max_scaled:average_distance_scaled")
    
    model = smf.mixedlm(formula, data, groups=data['ResponseId'])
    results = model.fit()
    
    print(f"\nMixed Effects Model Results for {trait}:")
    print(results.summary())
    
    return results

# 効果分析
def analyze_effects(results, data, trait):
    trait_effect = results.params[f'{trait}_distance_from_max_scaled']
    avg_distance_effect = results.params['average_distance_scaled']
    interaction_effect = results.params[f'{trait}_distance_from_max_scaled:average_distance_scaled']
    
    print(f"\nEffect Analysis for {trait}:")
    print(f"Effect of {trait} distance from max: {trait_effect}")
    print(f"Effect of average distance: {avg_distance_effect}")
    print(f"Interaction effect: {interaction_effect}")
    
    # 統計的検定
    for effect in [f'{trait}_distance_from_max_scaled', 'average_distance_scaled', f'{trait}_distance_from_max_scaled:average_distance_scaled']:
        t_stat = results.tvalues[effect]
        p_value = results.pvalues[effect]
        print(f"\n{effect}:")
        print(f"T-statistic: {t_stat}")
        print(f"P-value: {p_value}")

# メインの分析ループ
for trait in traits:
    # 混合効果モデル
    results = run_mixed_model(data, trait)
    
    # 効果分析
    analyze_effects(results, data, trait)
    
    # 多重共線性のチェック
    X = sm.add_constant(data[[f'{trait}_distance_from_max_scaled', 'average_distance_scaled']])
    vif = pd.DataFrame()
    vif["features"] = X.columns
    vif["VIF"] = [variance_inflation_factor(X.values, i) for i in range(X.shape[1])]
    print("\nVIF:\n", vif)

    # モデル診断（残差プロット）
    plt.figure(figsize=(10, 6))
    plt.scatter(results.fittedvalues, results.resid)
    plt.xlabel('Fitted values')
    plt.ylabel('Residuals')
    plt.title(f'Residual Plot for {trait}')
    plt.show()

    # 頑健性チェック（ブートストラップ）
    # Bootstrap manually since MixedLMResults does not have a bootstrap method
    """
    bootstrapped_params = []
    for _ in range(100):
        sample_indices = np.random.choice(data.index, size=len(data), replace=True)
        sample_data = data.loc[sample_indices]
        sample_results = run_mixed_model(sample_data, trait)
        bootstrapped_params.append(sample_results.params)
    
    bootstrapped_params_df = pd.DataFrame(bootstrapped_params)
    print("\nBootstrap Results:")
    print(bootstrapped_params_df.describe())
    """

Mixed Effects Model Results for EX:
                             Mixed Linear Model Regression Results
===============================================================================================
Model:                          MixedLM              Dependent Variable:              Ads_EX_1 
No. Observations:               90                   Method:                          REML     
No. Groups:                     90                   Scale:                           515.1420 
Min. group size:                1                    Log-Likelihood:                  -429.2133
Max. group size:                1                    Converged:                       Yes      
Mean group size:                1.0                                                            
-----------------------------------------------------------------------------------------------
                                                     Coef.  Std.Err.   z    P>|z| [0.025 0.975]
-----------------------------------------------------------------------------------------------
Intercept                                            53.719    3.210 16.733 0.000 47.427 60.011
EX_distance_from_max_scaled                           9.237    3.399  2.717 0.007  2.575 15.899
average_distance_scaled                               6.973    2.922  2.386 0.017  1.246 12.700
EX_distance_from_max_scaled:average_distance_scaled  -2.614    1.744 -1.499 0.134 -6.033  0.805
Group Var                                           515.142                                    
===============================================================================================


Effect Analysis for EX:
Effect of EX distance from max: 9.236731444764649
Effect of average distance: 6.9732976362044505
Interaction effect: -2.614142386564956

EX_distance_from_max_scaled:
T-statistic: 2.7174929666622285
P-value: 0.006577855342355496

average_distance_scaled:
T-statistic: 2.38647654804269
P-value: 0.01701069228185625

EX_distance_from_max_scaled:average_distance_scaled:
T-statistic: -1.4986328751579314
P-value: 0.1339688992290788

VIF:
                       features      VIF
0                        const  1.00000
1  EX_distance_from_max_scaled  1.02093
2      average_distance_scaled  1.02093

Mixed Effects Model Results for OP:
                             Mixed Linear Model Regression Results
===============================================================================================
Model:                          MixedLM              Dependent Variable:              Ads_OP_1 
No. Observations:               90                   Method:                          REML     
No. Groups:                     90                   Scale:                           491.0018 
Min. group size:                1                    Log-Likelihood:                  -427.1946
Max. group size:                1                    Converged:                       Yes      
Mean group size:                1.0                                                            
-----------------------------------------------------------------------------------------------
                                                     Coef.  Std.Err.   z    P>|z| [0.025 0.975]
-----------------------------------------------------------------------------------------------
Intercept                                            42.394    3.382 12.535 0.000 35.765 49.023
OP_distance_from_max_scaled                          -0.653    2.575 -0.254 0.800 -5.701  4.395
average_distance_scaled                               0.891    3.374  0.264 0.792 -5.722  7.503
OP_distance_from_max_scaled:average_distance_scaled   7.466    1.845  4.046 0.000  3.849 11.082
Group Var                                           491.002                                    
===============================================================================================


Effect Analysis for OP:
Effect of OP distance from max: -0.6529035119332733
Effect of average distance: 0.8906779078620093
Interaction effect: 7.465531880451053

OP_distance_from_max_scaled:
T-statistic: -0.25351162272480904
P-value: 0.7998728808201359

average_distance_scaled:
T-statistic: 0.26399905395088863
P-value: 0.7917806477377052

OP_distance_from_max_scaled:average_distance_scaled:
T-statistic: 4.0463938172233815
P-value: 5.201270142719014e-05

VIF:
                       features       VIF
0                        const  1.000000
1  OP_distance_from_max_scaled  1.045077
2      average_distance_scaled  1.045077

Mixed Effects Model Results for CO:
                             Mixed Linear Model Regression Results
===============================================================================================
Model:                          MixedLM              Dependent Variable:              Ads_CO_1 
No. Observations:               90                   Method:                          REML     
No. Groups:                     90                   Scale:                           495.4618 
Min. group size:                1                    Log-Likelihood:                  -427.6217
Max. group size:                1                    Converged:                       Yes      
Mean group size:                1.0                                                            
-----------------------------------------------------------------------------------------------
                                                     Coef.  Std.Err.   z    P>|z| [0.025 0.975]
-----------------------------------------------------------------------------------------------
Intercept                                            48.525    3.345 14.508 0.000 41.969 55.080
CO_distance_from_max_scaled                           3.057    2.611  1.171 0.242 -2.060  8.174
average_distance_scaled                               0.034    3.226  0.010 0.992 -6.289  6.357
CO_distance_from_max_scaled:average_distance_scaled  -2.201    1.338 -1.645 0.100 -4.824  0.422
Group Var                                           495.462                                    
===============================================================================================


Effect Analysis for CO:
Effect of CO distance from max: 3.057025850576481
Effect of average distance: 0.03356030426577199
Interaction effect: -2.2009905912533494

CO_distance_from_max_scaled:
T-statistic: 1.1708466648544527
P-value: 0.24166041671594096

average_distance_scaled:
T-statistic: 0.010402735306400398
P-value: 0.9916999678098024

CO_distance_from_max_scaled:average_distance_scaled:
T-statistic: -1.644863714962938
P-value: 0.09999791915021902

VIF:
                       features       VIF
0                        const  1.000000
1  CO_distance_from_max_scaled  1.028957
2      average_distance_scaled  1.028957

Mixed Effects Model Results for AG:
                             Mixed Linear Model Regression Results
===============================================================================================
Model:                          MixedLM              Dependent Variable:              Ads_AG_1 
No. Observations:               90                   Method:                          REML     
No. Groups:                     90                   Scale:                           358.9548 
Min. group size:                1                    Log-Likelihood:                  -413.8725
Max. group size:                1                    Converged:                       Yes      
Mean group size:                1.0                                                            
-----------------------------------------------------------------------------------------------
                                                     Coef.  Std.Err.   z    P>|z| [0.025 0.975]
-----------------------------------------------------------------------------------------------
Intercept                                            59.208    2.399 24.682 0.000 54.507 63.910
AG_distance_from_max_scaled                          -0.496    3.097 -0.160 0.873 -6.567  5.574
average_distance_scaled                               0.930    2.887  0.322 0.747 -4.729  6.589
AG_distance_from_max_scaled:average_distance_scaled   0.290    1.812  0.160 0.873 -3.262  3.842
Group Var                                           358.955                                    
===============================================================================================


Effect Analysis for AG:
Effect of AG distance from max: -0.496349475215326
Effect of average distance: 0.9300928988487126
Interaction effect: 0.2900427887798393

AG_distance_from_max_scaled:
T-statistic: -0.16025380699059646
P-value: 0.8726811451916974

average_distance_scaled:
T-statistic: 0.3221343180987998
P-value: 0.747350940744358

AG_distance_from_max_scaled:average_distance_scaled:
T-statistic: 0.16004444586776856
P-value: 0.8728460626994023

VIF:
                       features       VIF
0                        const  1.000000
1  AG_distance_from_max_scaled  1.041976
2      average_distance_scaled  1.041976

sandbox for analysis methods

merged_df

	ResponseId	OP_distance	OP_distance_from_max	CO_distance	CO_distance_from_max	EX_distance	EX_distance_from_max	AG_distance	AG_distance_from_max	total_distance	average_distance	Ads_OP_1	Ads_OP_2	Ads_OP_3	Ads_OP_4	Ads_OP_5	Ads_OP_6	Ads_CO_1	Ads_CO_2	Ads_CO_3	Ads_CO_4	Ads_CO_5	Ads_CO_6	Ads_EX_1	Ads_EX_2	Ads_EX_3	Ads_EX_4	Ads_EX_5	Ads_EX_6	Ads_AG_1	Ads_AG_2	Ads_AG_3	Ads_AG_4	Ads_AG_5	Ads_AG_6	Desc_OP_1	Desc_CO_1	Desc_EX_1	Desc_AG_1	EX_distance_from_max_scaled	OP_distance_from_max_scaled	CO_distance_from_max_scaled	AG_distance_from_max_scaled	average_distance_scaled
0	641357b26802e397b5627d5e	0.333333	0.833333	0.166667	0.833333	0.666667	1.166667	0.833333	1.833333	2.000000	0.500000	89	75	89	84	71	75	81	66	74	77	79	82	72	68	67	68	76	69	71	78	70	68	85	84	63	50	53	50	-0.881537	-0.550930	-0.277944	1.076488	-0.484285
1	66294a585f1cf3fdaeb80120	0.666667	1.833333	0.666667	0.333333	1.000000	0.833333	0.833333	1.833333	3.166667	0.791667	72	50	53	28	50	59	34	50	50	59	46	21	54	50	42	38	45	62	56	50	55	64	58	60	91	10	80	60	-1.271406	0.820048	-0.954025	1.076488	0.702872
2	661571d6cc16ee34676734df	0.333333	0.833333	1.000000	2.000000	0.000000	1.833333	0.666667	1.666667	2.000000	0.500000	29	49	37	45	34	31	60	60	58	59	62	80	61	100	56	57	61	67	65	59	59	59	65	71	28	49	57	57	-0.101799	-0.550930	1.299577	0.837857	-0.484285
3	6629f95a14b6006e6a4b0aef	0.333333	0.833333	0.500000	1.500000	0.166667	2.000000	0.333333	0.666667	1.333333	0.333333	37	37	40	42	40	41	64	66	71	60	66	66	33	34	38	35	36	35	64	62	59	60	62	66	68	62	65	36	0.093135	-0.550930	0.623496	-0.593924	-1.162660
4	66463d8b1984a328f78252d7	1.166667	0.000000	1.666667	2.666667	0.000000	1.833333	0.333333	1.333333	3.166667	0.791667	13	39	33	23	40	15	81	73	77	85	61	84	16	21	20	30	14	26	20	40	8	28	18	16	14	94	65	9	-0.101799	-1.693412	2.201018	0.360597	0.702872
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
85	5d4a6459061e2a000138bad3	1.166667	0.000000	0.666667	0.333333	0.500000	1.333333	0.333333	0.666667	2.666667	0.666667	48	73	48	48	48	47	11	40	10	10	9	10	93	52	96	95	95	95	14	80	14	46	17	44	7	26	54	47	-0.686603	-1.693412	-0.954025	-0.593924	0.194091
86	62b473736986d2b27c1427d2	1.166667	2.333333	0.333333	1.333333	1.000000	0.833333	0.500000	1.500000	3.000000	0.750000	72	27	27	40	17	30	71	88	26	26	68	82	33	15	76	30	70	13	74	26	66	25	68	45	21	28	84	68	-1.271406	1.505537	0.398136	0.599227	0.533278
87	5f3ac1732efa0a74f975b1a8	0.833333	2.000000	0.333333	0.666667	0.333333	1.500000	0.833333	0.166667	2.333333	0.583333	5	14	0	3	31	0	24	0	26	0	46	0	29	1	3	50	28	23	81	58	77	60	51	59	100	59	0	11	-0.491668	1.048544	-0.503304	-1.309815	-0.145097
88	6601b51675d287d6a62f11c4	0.833333	0.333333	0.333333	0.666667	0.500000	1.333333	0.833333	0.166667	2.500000	0.625000	18	44	19	23	20	24	33	45	34	40	39	35	18	40	17	17	20	22	71	83	79	75	82	89	45	44	83	57	-0.686603	-1.236419	-0.503304	-1.309815	0.024497
89	6658822c0f5b1367a1585ee1	0.666667	1.833333	1.000000	0.000000	1.333333	0.500000	0.666667	0.333333	3.666667	0.916667	40	50	40	40	40	40	80	80	80	85	90	90	90	50	90	80	100	90	60	70	60	50	70	70	40	70	60	70	-1.661275	0.820048	-1.404745	-1.071185	1.211654

90 rows × 44 columns

import pandas as pd
import numpy as np
import statsmodels.api as sm
import statsmodels.formula.api as smf
from scipy import stats
from sklearn.preprocessing import StandardScaler
from sklearn.mixture import GaussianMixture
import matplotlib.pyplot as plt
import seaborn as sns

data = merged_df

# a. 差分スコアを用いた分析
"""
この分析は、Blended Personality広告と特性特化広告の直接比較を可能にします。
正の差分スコアは、Blended Personality広告がより好まれることを示し、
特性からの距離や平均距離との関係を明らかにすることで、どのような条件下でBlended Personality広告が
より効果的かを理解するのに役立ちます。
"""
for trait in ['EX', 'OP', 'CO', 'AG']:
    # Blended Personality広告の平均スコアを計算
    blended_cols = [f'Ads_{trait}_{i}' for i in range(1, 7)]
    data[f'{trait}_blended_score'] = data[blended_cols].mean(axis=1)
    
    # 差分スコアの計算（Blended - Trait Specific）
    data[f'{trait}_diff_score'] = data[f'{trait}_blended_score'] - data[f'Ads_{trait}_1']

def analyze_diff_score(trait):
    model = smf.ols(f'{trait}_diff_score ~ {trait}_distance_from_max_scaled + average_distance_scaled', data=data)
    results = model.fit()
    print(f"\nResults for {trait}:")
    print(results.summary())

for trait in ['EX', 'OP', 'CO', 'AG']:
    analyze_diff_score(trait)

Results for EX:
                            OLS Regression Results                            
==============================================================================
Dep. Variable:          EX_diff_score   R-squared:                       0.006
Model:                            OLS   Adj. R-squared:                 -0.017
Method:                 Least Squares   F-statistic:                    0.2507
Date:                Thu, 25 Jul 2024   Prob (F-statistic):              0.779
Time:                        07:59:38   Log-Likelihood:                -315.81
No. Observations:                  90   AIC:                             637.6
Df Residuals:                      87   BIC:                             645.1
Df Model:                           2                                         
Covariance Type:            nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -0.9759      0.867     -1.126      0.263      -2.699       0.747
EX_distance_from_max_scaled    -0.6202      0.876     -0.708      0.481      -2.361       1.121
average_distance_scaled         0.0862      0.876      0.098      0.922      -1.655       1.827
==============================================================================
Omnibus:                       50.893   Durbin-Watson:                   2.169
Prob(Omnibus):                  0.000   Jarque-Bera (JB):              264.181
Skew:                          -1.674   Prob(JB):                     4.30e-58
Kurtosis:                      10.697   Cond. No.                         1.16
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for OP:
                            OLS Regression Results                            
==============================================================================
Dep. Variable:          OP_diff_score   R-squared:                       0.008
Model:                            OLS   Adj. R-squared:                 -0.015
Method:                 Least Squares   F-statistic:                    0.3423
Date:                Thu, 25 Jul 2024   Prob (F-statistic):              0.711
Time:                        07:59:38   Log-Likelihood:                -376.71
No. Observations:                  90   AIC:                             759.4
Df Residuals:                      87   BIC:                             766.9
Df Model:                           2                                         
Covariance Type:            nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -0.9389      1.705     -0.551      0.583      -4.328       2.451
OP_distance_from_max_scaled    -1.4241      1.743     -0.817      0.416      -4.889       2.041
average_distance_scaled         0.5196      1.743      0.298      0.766      -2.945       3.985
==============================================================================
Omnibus:                       26.642   Durbin-Watson:                   2.470
Prob(Omnibus):                  0.000   Jarque-Bera (JB):              269.685
Skew:                          -0.254   Prob(JB):                     2.75e-59
Kurtosis:                      11.465   Cond. No.                         1.23
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for CO:
                            OLS Regression Results                            
==============================================================================
Dep. Variable:          CO_diff_score   R-squared:                       0.008
Model:                            OLS   Adj. R-squared:                 -0.015
Method:                 Least Squares   F-statistic:                    0.3621
Date:                Thu, 25 Jul 2024   Prob (F-statistic):              0.697
Time:                        07:59:38   Log-Likelihood:                -352.74
No. Observations:                  90   AIC:                             711.5
Df Residuals:                      87   BIC:                             719.0
Df Model:                           2                                         
Covariance Type:            nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -0.5667      1.307     -0.434      0.666      -3.164       2.030
CO_distance_from_max_scaled    -1.0147      1.325     -0.766      0.446      -3.649       1.620
average_distance_scaled         0.6560      1.325      0.495      0.622      -1.978       3.290
==============================================================================
Omnibus:                       34.490   Durbin-Watson:                   2.404
Prob(Omnibus):                  0.000   Jarque-Bera (JB):              615.522
Skew:                           0.359   Prob(JB):                    2.19e-134
Kurtosis:                      15.792   Cond. No.                         1.18
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for AG:
                            OLS Regression Results                            
==============================================================================
Dep. Variable:          AG_diff_score   R-squared:                       0.025
Model:                            OLS   Adj. R-squared:                  0.003
Method:                 Least Squares   F-statistic:                     1.130
Date:                Thu, 25 Jul 2024   Prob (F-statistic):              0.328
Time:                        07:59:38   Log-Likelihood:                -327.21
No. Observations:                  90   AIC:                             660.4
Df Residuals:                      87   BIC:                             667.9
Df Model:                           2                                         
Covariance Type:            nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -0.0574      0.984     -0.058      0.954      -2.013       1.898
AG_distance_from_max_scaled     0.3739      1.004      0.372      0.711      -1.622       2.370
average_distance_scaled        -1.5083      1.004     -1.502      0.137      -3.505       0.488
==============================================================================
Omnibus:                       73.644   Durbin-Watson:                   2.038
Prob(Omnibus):                  0.000   Jarque-Bera (JB):              592.569
Skew:                          -2.491   Prob(JB):                    2.11e-129
Kurtosis:                      14.541   Cond. No.                         1.23
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

"""
マルチレベルモデルは、個人内での広告タイプの効果と個人間での特性の効果を同時に分析します。
これにより、Blended Personality広告が全体的にどの程度好まれるか、
また特定の特性や条件下でどのように効果が変化するかを理解できます。
"""
# b. マルチレベルモデル
data_long = pd.melt(data, 
                    id_vars=['ResponseId', 'EX_distance_from_max_scaled', 'OP_distance_from_max_scaled', 
                             'CO_distance_from_max_scaled', 'AG_distance_from_max_scaled', 'average_distance_scaled'],
                    value_vars=[col for col in data.columns if col.startswith('Ads_') and col.endswith(('1', '2', '3', '4', '5', '6'))],
                    var_name='ad_type', value_name='preference_score')

data_long['is_blended'] = ~data_long['ad_type'].str.endswith('1')
data_long['trait'] = data_long['ad_type'].str.split('_').str[1]
data_long['question_type'] = data_long['ad_type'].str.split('_').str[2]

model = smf.mixedlm("preference_score ~ is_blended + trait + question_type + is_blended:trait + average_distance_scaled", 
                    data=data_long, groups=data_long["ResponseId"])
results = model.fit()
print(results.summary())

                            Mixed Linear Model Regression Results
==============================================================================================
Model:                        MixedLM           Dependent Variable:           preference_score
No. Observations:             2160              Method:                       REML            
No. Groups:                   90                Scale:                        703.6243        
Min. group size:              24                Log-Likelihood:               -10197.7585     
Max. group size:              24                Converged:                    No              
Mean group size:              24.0                                                            
----------------------------------------------------------------------------------------------
                                Coef.     Std.Err.     z    P>|z|     [0.025         0.975]   
----------------------------------------------------------------------------------------------
Intercept                        59.267        3.234 18.328 0.000         52.929        65.605
is_blended[T.True]              629.051 54478103.123  0.000 1.000 -106774491.016 106775749.117
trait[T.CO]                     -11.111        3.954 -2.810 0.005        -18.861        -3.361
trait[T.EX]                      -5.922        3.954 -1.498 0.134        -13.672         1.828
trait[T.OP]                     -15.322        3.954 -3.875 0.000        -23.072        -7.572
question_type[T.2]             -627.599 54478103.123 -0.000 1.000 -106775747.666 106774492.467
question_type[T.3]             -629.122 54478103.123 -0.000 1.000 -106775749.188 106774490.945
question_type[T.4]             -630.369 54478103.123 -0.000 1.000 -106775750.435 106774489.698
question_type[T.5]             -629.324 54478103.123 -0.000 1.000 -106775749.391 106774490.742
question_type[T.6]             -629.183 54478103.123 -0.000 1.000 -106775749.249 106774490.884
is_blended[T.True]:trait[T.CO]   -0.611        4.332 -0.141 0.888         -9.101         7.879
is_blended[T.True]:trait[T.EX]   -1.102        4.332 -0.254 0.799         -9.592         7.388
is_blended[T.True]:trait[T.OP]   -1.058        4.332 -0.244 0.807         -9.548         7.432
average_distance_scaled           2.150        1.722  1.249 0.212         -1.225         5.525
Group Var                       237.516        1.897                                          
==============================================================================================

"""
潜在クラス分析は、回答者を類似した特性と広告選好を持つグループに分類します。
これにより、Blended Personality広告がどのようなタイプの人々により効果的であるかを
探索的に理解することができ、GPT-4の広告生成能力の強みと限界を明らかにするのに役立ちます。
"""
# c. 潜在クラス分析（ここではGaussian Mixture Modelを使用）
features = ['EX_distance_from_max_scaled', 'OP_distance_from_max_scaled', 
            'CO_distance_from_max_scaled', 'AG_distance_from_max_scaled', 
            'average_distance_scaled'] + [f'{trait}_diff_score' for trait in ['EX', 'OP', 'CO', 'AG']]

X = data[features]
X_scaled = StandardScaler().fit_transform(X)

# モデルの適合（クラス数は仮に3としています）
gmm = GaussianMixture(n_components=3, random_state=42)
data['cluster'] = gmm.fit_predict(X_scaled)

# クラスタごとの特徴を可視化
plt.figure(figsize=(12, 6))
sns.boxplot(x='cluster', y='value', hue='variable', 
            data=pd.melt(data[features + ['cluster']], id_vars='cluster'))
plt.title('Feature Distribution by Cluster')
plt.show()

"""
閾値を用いた分析は、Blended Personality広告の効果が特定の条件下で
どのように変化するかを明らかにします。これは、GPT-4が生成する広告の
効果が、ターゲットとなる個人の特性プロファイルにどの程度依存するかを
理解するのに役立ちます。
"""

# d. 閾値を用いた分析
data['high_average_distance'] = data['average_distance_scaled'] > data['average_distance_scaled'].median()

for trait in ['EX', 'OP', 'CO', 'AG']:
    t_stat, p_value = stats.ttest_ind(
        data[data['high_average_distance']][f'{trait}_diff_score'],
        data[~data['high_average_distance']][f'{trait}_diff_score']
    )
    print(f"\n{trait} - T-test for diff_score between high and low average_distance groups:")
    print(f"T-statistic: {t_stat}, P-value: {p_value}")

EX - T-test for diff_score between high and low average_distance groups:
T-statistic: -0.060950428225559815, P-value: 0.9515369015866464

OP - T-test for diff_score between high and low average_distance groups:
T-statistic: -0.19554658783863332, P-value: 0.8454158847783683

CO - T-test for diff_score between high and low average_distance groups:
T-statistic: 0.48716244673001935, P-value: 0.6273545341037027

AG - T-test for diff_score between high and low average_distance groups:
T-statistic: -1.1120780933553684, P-value: 0.2691334656935769

"""
非線形関係の分析は、特性からの距離や平均距離と広告の好みの間の複雑な関係を
捉えることができます。これにより、GPT-4が生成する広告の効果が
線形的に変化するのではなく、特定の閾値や範囲で急激に変化する可能性を
検討することができます。
"""
# e. 非線形関係の考慮
for trait in ['EX', 'OP', 'CO', 'AG']:
    model = smf.ols(f'{trait}_diff_score ~ {trait}_distance_from_max_scaled + I({trait}_distance_from_max_scaled**2) + '
                    'average_distance_scaled + I(average_distance_scaled**2)', data=data)
    results = model.fit()
    print(f"\nNon-linear model results for {trait}:")
    print(results.summary())

Non-linear model results for EX:
                            OLS Regression Results                            
==============================================================================
Dep. Variable:          EX_diff_score   R-squared:                       0.033
Model:                            OLS   Adj. R-squared:                 -0.012
Method:                 Least Squares   F-statistic:                    0.7303
Date:                Thu, 25 Jul 2024   Prob (F-statistic):              0.574
Time:                        07:59:45   Log-Likelihood:                -314.55
No. Observations:                  90   AIC:                             639.1
Df Residuals:                      85   BIC:                             651.6
Df Model:                           4                                         
Covariance Type:            nonrobust                                         
=======================================================================================================
                                          coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------------------
Intercept                               0.3164      1.432      0.221      0.826      -2.531       3.164
EX_distance_from_max_scaled            -0.5842      0.887     -0.659      0.512      -2.348       1.180
I(EX_distance_from_max_scaled ** 2)    -1.4240      0.922     -1.544      0.126      -3.258       0.410
average_distance_scaled                 0.9824      1.095      0.897      0.372      -1.195       3.160
I(average_distance_scaled ** 2)         0.1316      0.678      0.194      0.847      -1.217       1.480
==============================================================================
Omnibus:                       40.873   Durbin-Watson:                   2.194
Prob(Omnibus):                  0.000   Jarque-Bera (JB):              170.764
Skew:                          -1.353   Prob(JB):                     8.30e-38
Kurtosis:                       9.181   Cond. No.                         4.33
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Non-linear model results for OP:
                            OLS Regression Results                            
==============================================================================
Dep. Variable:          OP_diff_score   R-squared:                       0.029
Model:                            OLS   Adj. R-squared:                 -0.016
Method:                 Least Squares   F-statistic:                    0.6415
Date:                Thu, 25 Jul 2024   Prob (F-statistic):              0.634
Time:                        07:59:45   Log-Likelihood:                -375.72
No. Observations:                  90   AIC:                             761.4
Df Residuals:                      85   BIC:                             773.9
Df Model:                           4                                         
Covariance Type:            nonrobust                                         
=======================================================================================================
                                          coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------------------
Intercept                               1.3075      2.425      0.539      0.591      -3.515       6.130
OP_distance_from_max_scaled            -0.7876      1.850     -0.426      0.671      -4.467       2.892
I(OP_distance_from_max_scaled ** 2)    -1.4370      1.234     -1.165      0.247      -3.890       1.016
average_distance_scaled                 1.6995      1.948      0.872      0.385      -2.174       5.573
I(average_distance_scaled ** 2)        -0.8095      1.328     -0.610      0.544      -3.449       1.830
==============================================================================
Omnibus:                       27.763   Durbin-Watson:                   2.471
Prob(Omnibus):                  0.000   Jarque-Bera (JB):              279.596
Skew:                          -0.343   Prob(JB):                     1.93e-61
Kurtosis:                      11.607   Cond. No.                         3.74
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Non-linear model results for CO:
                            OLS Regression Results                            
==============================================================================
Dep. Variable:          CO_diff_score   R-squared:                       0.011
Model:                            OLS   Adj. R-squared:                 -0.036
Method:                 Least Squares   F-statistic:                    0.2368
Date:                Thu, 25 Jul 2024   Prob (F-statistic):              0.917
Time:                        07:59:45   Log-Likelihood:                -352.62
No. Observations:                  90   AIC:                             715.2
Df Residuals:                      85   BIC:                             727.7
Df Model:                           4                                         
Covariance Type:            nonrobust                                         
=======================================================================================================
                                          coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------------------
Intercept                              -0.2973      2.103     -0.141      0.888      -4.478       3.884
CO_distance_from_max_scaled            -1.0020      1.441     -0.695      0.489      -3.868       1.864
I(CO_distance_from_max_scaled ** 2)     0.2177      1.346      0.162      0.872      -2.459       2.895
average_distance_scaled                 0.7036      1.600      0.440      0.661      -2.477       3.884
I(average_distance_scaled ** 2)        -0.4871      1.035     -0.471      0.639      -2.545       1.571
==============================================================================
Omnibus:                       34.408   Durbin-Watson:                   2.406
Prob(Omnibus):                  0.000   Jarque-Bera (JB):              607.876
Skew:                           0.361   Prob(JB):                    1.00e-132
Kurtosis:                      15.711   Cond. No.                         4.26
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Non-linear model results for AG:
                            OLS Regression Results                            
==============================================================================
Dep. Variable:          AG_diff_score   R-squared:                       0.026
Model:                            OLS   Adj. R-squared:                 -0.020
Method:                 Least Squares   F-statistic:                    0.5576
Date:                Thu, 25 Jul 2024   Prob (F-statistic):              0.694
Time:                        07:59:45   Log-Likelihood:                -327.20
No. Observations:                  90   AIC:                             664.4
Df Residuals:                      85   BIC:                             676.9
Df Model:                           4                                         
Covariance Type:            nonrobust                                         
=======================================================================================================
                                          coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------------------
Intercept                              -0.1085      1.499     -0.072      0.942      -3.090       2.873
AG_distance_from_max_scaled             0.3610      1.049      0.344      0.732      -1.725       2.447
I(AG_distance_from_max_scaled ** 2)    -0.0640      0.986     -0.065      0.948      -2.025       1.897
average_distance_scaled                -1.5156      1.145     -1.324      0.189      -3.792       0.761
I(average_distance_scaled ** 2)         0.1151      0.804      0.143      0.886      -1.483       1.713
==============================================================================
Omnibus:                       73.405   Durbin-Watson:                   2.043
Prob(Omnibus):                  0.000   Jarque-Bera (JB):              584.951
Skew:                          -2.485   Prob(JB):                    9.54e-128
Kurtosis:                      14.458   Cond. No.                         3.89
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

sandbox 2

import pandas as pd
import numpy as np
import statsmodels.api as sm
import statsmodels.formula.api as smf
from scipy import stats
from sklearn.preprocessing import StandardScaler
from sklearn.cluster import KMeans
import matplotlib.pyplot as plt
import seaborn as sns

# データの読み込み（既存のデータフレームを 'data' とします）
data = merged_df

# 1. 相対的選好スコアの作成
for trait in ['EX', 'OP', 'CO', 'AG']:
    for i in range(1, 7):
        data[f'{trait}_relative_score_{i}'] = data[f'Ads_{trait}_{i}'] - 50

# 2. 特性別分析
def analyze_trait(trait):
    for i in range(1, 7):
        model = smf.ols(f'{trait}_relative_score_{i} ~ {trait}_distance_from_max_scaled + average_distance_scaled', data=data)
        results = model.fit()
        print(f"\nResults for {trait}, question {i}:")
        print(results.summary())

for trait in ['EX', 'OP', 'CO', 'AG']:
    analyze_trait(trait)

Results for EX, question 1:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_1   R-squared:                       0.103
Model:                             OLS   Adj. R-squared:                  0.082
Method:                  Least Squares   F-statistic:                     4.986
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):            0.00891
Time:                         07:59:45   Log-Likelihood:                -438.09
No. Observations:                   90   AIC:                             882.2
Df Residuals:                       87   BIC:                             889.7
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                       3.3444      3.373      0.992      0.324      -3.360      10.049
EX_distance_from_max_scaled     7.5517      3.408      2.216      0.029       0.778      14.326
average_distance_scaled         6.5080      3.408      1.910      0.059      -0.266      13.282
==============================================================================
Omnibus:                       21.498   Durbin-Watson:                   1.739
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                4.968
Skew:                          -0.102   Prob(JB):                       0.0834
Kurtosis:                       1.867   Cond. No.                         1.16
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for EX, question 2:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_2   R-squared:                       0.110
Model:                             OLS   Adj. R-squared:                  0.090
Method:                  Least Squares   F-statistic:                     5.384
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):            0.00624
Time:                         07:59:45   Log-Likelihood:                -428.58
No. Observations:                   90   AIC:                             863.2
Df Residuals:                       87   BIC:                             870.7
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                       1.1111      3.035      0.366      0.715      -4.921       7.143
EX_distance_from_max_scaled     7.8445      3.066      2.558      0.012       1.750      13.939
average_distance_scaled         5.1130      3.066      1.668      0.099      -0.981      11.208
==============================================================================
Omnibus:                        9.881   Durbin-Watson:                   1.766
Prob(Omnibus):                  0.007   Jarque-Bera (JB):                3.406
Skew:                           0.022   Prob(JB):                        0.182
Kurtosis:                       2.048   Cond. No.                         1.16
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for EX, question 3:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_3   R-squared:                       0.071
Model:                             OLS   Adj. R-squared:                  0.049
Method:                  Least Squares   F-statistic:                     3.312
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):             0.0411
Time:                         07:59:45   Log-Likelihood:                -441.95
No. Observations:                   90   AIC:                             889.9
Df Residuals:                       87   BIC:                             897.4
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                       2.7778      3.521      0.789      0.432      -4.220       9.776
EX_distance_from_max_scaled     4.2020      3.557      1.181      0.241      -2.869      11.273
average_distance_scaled         7.4486      3.557      2.094      0.039       0.378      14.519
==============================================================================
Omnibus:                       66.369   Durbin-Watson:                   1.782
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.181
Skew:                          -0.103   Prob(JB):                       0.0276
Kurtosis:                       1.632   Cond. No.                         1.16
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for EX, question 4:
Results for EX, question 4:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_4   R-squared:                       0.096
Model:                             OLS   Adj. R-squared:                  0.075
Method:                  Least Squares   F-statistic:                     4.631
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):             0.0123
Time:                         07:59:45   Log-Likelihood:                -435.15
No. Observations:                   90   AIC:                             876.3
Df Residuals:                       87   BIC:                             883.8
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                       0.6222      3.264      0.191      0.849      -5.866       7.111
EX_distance_from_max_scaled     7.0139      3.298      2.126      0.036       0.458      13.570
average_distance_scaled         6.1038      3.298      1.850      0.068      -0.452      12.660
==============================================================================
Omnibus:                       25.099   Durbin-Watson:                   1.671
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.164
Skew:                           0.047   Prob(JB):                       0.0756
Kurtosis:                       1.830   Cond. No.                         1.16
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for EX, question 5:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_5   R-squared:                       0.114
Model:                             OLS   Adj. R-squared:                  0.094
Method:                  Least Squares   F-statistic:                     5.592
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):            0.00519
Time:                         07:59:45   Log-Likelihood:                -439.57
No. Observations:                   90   AIC:                             885.1
Df Residuals:                       87   BIC:                             892.6
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                       3.3222      3.429      0.969      0.335      -3.493      10.137
EX_distance_from_max_scaled     6.9460      3.464      2.005      0.048       0.060      13.832
average_distance_scaled         8.1827      3.464      2.362      0.020       1.297      15.069
==============================================================================
Omnibus:                       24.032   Durbin-Watson:                   1.687
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.290
Skew:                          -0.133   Prob(JB):                       0.0710
Kurtosis:                       1.843   Cond. No.                         1.16
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for EX, question 6:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_6   R-squared:                       0.096
Model:                             OLS   Adj. R-squared:                  0.075
Method:                  Least Squares   F-statistic:                     4.612
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):             0.0125
Time:                         07:59:45   Log-Likelihood:                -443.11
No. Observations:                   90   AIC:                             892.2
Df Residuals:                       87   BIC:                             899.7
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                       3.0333      3.567      0.850      0.397      -4.056      10.122
EX_distance_from_max_scaled     8.0312      3.604      2.229      0.028       0.868      15.194
average_distance_scaled         6.2094      3.604      1.723      0.088      -0.953      13.372
==============================================================================
Omnibus:                       27.502   Durbin-Watson:                   1.736
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.457
Skew:                          -0.101   Prob(JB):                       0.0653
Kurtosis:                       1.811   Cond. No.                         1.16
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for OP, question 1:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_1   R-squared:                       0.022
Model:                             OLS   Adj. R-squared:                 -0.001
Method:                  Least Squares   F-statistic:                    0.9581
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.388
Time:                         07:59:45   Log-Likelihood:                -437.98
No. Observations:                   90   AIC:                             882.0
Df Residuals:                       87   BIC:                             889.5
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -6.0556      3.369     -1.798      0.076     -12.751       0.640
OP_distance_from_max_scaled     4.6903      3.444      1.362      0.177      -2.155      11.535
average_distance_scaled        -0.1388      3.444     -0.040      0.968      -6.984       6.706
==============================================================================
Omnibus:                       38.063   Durbin-Watson:                   1.888
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                6.967
Skew:                           0.267   Prob(JB):                       0.0307
Kurtosis:                       1.746   Cond. No.                         1.23
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for OP, question 2:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_2   R-squared:                       0.007
Model:                             OLS   Adj. R-squared:                 -0.016
Method:                  Least Squares   F-statistic:                    0.2889
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.750
Time:                         07:59:45   Log-Likelihood:                -421.82
No. Observations:                   90   AIC:                             849.6
Df Residuals:                       87   BIC:                             857.1
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -4.2111      2.815     -1.496      0.138      -9.806       1.384
OP_distance_from_max_scaled     2.1416      2.878      0.744      0.459      -3.578       7.862
average_distance_scaled        -0.0090      2.878     -0.003      0.998      -5.729       5.711
==============================================================================
Omnibus:                        3.824   Durbin-Watson:                   2.223
Prob(Omnibus):                  0.148   Jarque-Bera (JB):                2.089
Skew:                           0.064   Prob(JB):                        0.352
Kurtosis:                       2.265   Cond. No.                         1.23
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for OP, question 3:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_3   R-squared:                       0.008
Model:                             OLS   Adj. R-squared:                 -0.015
Method:                  Least Squares   F-statistic:                    0.3331
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.718
Time:                         07:59:45   Log-Likelihood:                -440.33
No. Observations:                   90   AIC:                             886.7
Df Residuals:                       87   BIC:                             894.2
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -6.9000      3.458     -1.995      0.049     -13.773      -0.027
OP_distance_from_max_scaled     2.8759      3.535      0.814      0.418      -4.151       9.902
average_distance_scaled        -0.8247      3.535     -0.233      0.816      -7.851       6.202
==============================================================================
Omnibus:                       46.816   Durbin-Watson:                   1.936
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.457
Skew:                           0.276   Prob(JB):                       0.0240
Kurtosis:                       1.702   Cond. No.                         1.23
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for OP, question 4:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_4   R-squared:                       0.022
Model:                             OLS   Adj. R-squared:                 -0.000
Method:                  Least Squares   F-statistic:                    0.9930
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.375
Time:                         07:59:45   Log-Likelihood:                -432.26
No. Observations:                   90   AIC:                             870.5
Df Residuals:                       87   BIC:                             878.0
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -8.5222      3.161     -2.696      0.008     -14.806      -2.239
OP_distance_from_max_scaled     3.8287      3.232      1.185      0.239      -2.595      10.252
average_distance_scaled         1.6172      3.232      0.500      0.618      -4.806       8.041
==============================================================================
Omnibus:                       13.810   Durbin-Watson:                   1.966
Prob(Omnibus):                  0.001   Jarque-Bera (JB):                5.697
Skew:                           0.367   Prob(JB):                       0.0579
Kurtosis:                       2.010   Cond. No.                         1.23
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for OP, question 5:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_5   R-squared:                       0.010
Model:                             OLS   Adj. R-squared:                 -0.013
Method:                  Least Squares   F-statistic:                    0.4416
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.644
Time:                         07:59:45   Log-Likelihood:                -435.37
No. Observations:                   90   AIC:                             876.7
Df Residuals:                       87   BIC:                             884.2
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -8.2333      3.273     -2.516      0.014     -14.738      -1.729
OP_distance_from_max_scaled     2.8814      3.345      0.861      0.391      -3.768       9.531
average_distance_scaled         0.6320      3.345      0.189      0.851      -6.017       7.281
==============================================================================
Omnibus:                       17.137   Durbin-Watson:                   2.016
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                6.616
Skew:                           0.413   Prob(JB):                       0.0366
Kurtosis:                       1.959   Cond. No.                         1.23
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for OP, question 6:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_6   R-squared:                       0.012
Model:                             OLS   Adj. R-squared:                 -0.011
Method:                  Least Squares   F-statistic:                    0.5291
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.591
Time:                         07:59:45   Log-Likelihood:                -439.62
No. Observations:                   90   AIC:                             885.2
Df Residuals:                       87   BIC:                             892.7
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -8.0444      3.431     -2.345      0.021     -14.863      -1.226
OP_distance_from_max_scaled     3.1791      3.507      0.906      0.367      -3.792      10.150
average_distance_scaled         1.0082      3.507      0.287      0.774      -5.962       7.979
==============================================================================
Omnibus:                       38.012   Durbin-Watson:                   1.871
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.430
Skew:                           0.323   Prob(JB):                       0.0244
Kurtosis:                       1.749   Cond. No.                         1.23
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for CO, question 1:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_1   R-squared:                       0.003
Model:                             OLS   Adj. R-squared:                 -0.020
Method:                  Least Squares   F-statistic:                    0.1193
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.888
Time:                         07:59:45   Log-Likelihood:                -436.31
No. Observations:                   90   AIC:                             878.6
Df Residuals:                       87   BIC:                             886.1
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -1.8444      3.307     -0.558      0.578      -8.417       4.728
CO_distance_from_max_scaled     1.6386      3.354      0.488      0.626      -5.029       8.306
average_distance_scaled        -0.2937      3.354     -0.088      0.930      -6.961       6.374
==============================================================================
Omnibus:                       40.136   Durbin-Watson:                   2.122
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                6.125
Skew:                           0.064   Prob(JB):                       0.0468
Kurtosis:                       1.728   Cond. No.                         1.18
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for CO, question 2:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_2   R-squared:                       0.005
Model:                             OLS   Adj. R-squared:                 -0.018
Method:                  Least Squares   F-statistic:                    0.1990
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.820
Time:                         07:59:45   Log-Likelihood:                -428.20
No. Observations:                   90   AIC:                             862.4
Df Residuals:                       87   BIC:                             869.9
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -0.8667      3.022     -0.287      0.775      -6.873       5.140
CO_distance_from_max_scaled    -0.8477      3.065     -0.277      0.783      -6.940       5.245
average_distance_scaled        -1.5712      3.065     -0.513      0.610      -7.664       4.521
==============================================================================
Omnibus:                        6.127   Durbin-Watson:                   2.068
Prob(Omnibus):                  0.047   Jarque-Bera (JB):                2.687
Skew:                           0.052   Prob(JB):                        0.261
Kurtosis:                       2.160   Cond. No.                         1.18
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for CO, question 3:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_3   R-squared:                       0.002
Model:                             OLS   Adj. R-squared:                 -0.021
Method:                  Least Squares   F-statistic:                   0.08013
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.923
Time:                         07:59:45   Log-Likelihood:                -439.72
No. Observations:                   90   AIC:                             885.4
Df Residuals:                       87   BIC:                             892.9
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -3.6444      3.435     -1.061      0.292     -10.471       3.182
CO_distance_from_max_scaled     0.3165      3.484      0.091      0.928      -6.608       7.241
average_distance_scaled         1.2860      3.484      0.369      0.713      -5.639       8.211
==============================================================================
Omnibus:                       73.154   Durbin-Watson:                   1.993
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.397
Skew:                           0.115   Prob(JB):                       0.0248
Kurtosis:                       1.615   Cond. No.                         1.18
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for CO, question 4:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_4   R-squared:                       0.003
Model:                             OLS   Adj. R-squared:                 -0.020
Method:                  Least Squares   F-statistic:                    0.1242
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.883
Time:                         07:59:45   Log-Likelihood:                -434.09
No. Observations:                   90   AIC:                             874.2
Df Residuals:                       87   BIC:                             881.7
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -2.8444      3.226     -0.882      0.380      -9.257       3.568
CO_distance_from_max_scaled     0.4840      3.273      0.148      0.883      -6.021       6.989
average_distance_scaled         1.4546      3.273      0.444      0.658      -5.050       7.959
==============================================================================
Omnibus:                       25.081   Durbin-Watson:                   2.014
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.246
Skew:                           0.091   Prob(JB):                       0.0726
Kurtosis:                       1.831   Cond. No.                         1.18
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for CO, question 5:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_5   R-squared:                       0.003
Model:                             OLS   Adj. R-squared:                 -0.020
Method:                  Least Squares   F-statistic:                    0.1145
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.892
Time:                         07:59:45   Log-Likelihood:                -438.00
No. Observations:                   90   AIC:                             882.0
Df Residuals:                       87   BIC:                             889.5
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -2.1556      3.370     -0.640      0.524      -8.853       4.542
CO_distance_from_max_scaled     0.8992      3.418      0.263      0.793      -5.895       7.693
average_distance_scaled         1.1964      3.418      0.350      0.727      -5.597       7.990
==============================================================================
Omnibus:                       32.478   Durbin-Watson:                   2.024
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.788
Skew:                           0.100   Prob(JB):                       0.0554
Kurtosis:                       1.774   Cond. No.                         1.18
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for CO, question 6:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_6   R-squared:                       0.001
Model:                             OLS   Adj. R-squared:                 -0.022
Method:                  Least Squares   F-statistic:                   0.06073
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.941
Time:                         07:59:45   Log-Likelihood:                -445.36
No. Observations:                   90   AIC:                             896.7
Df Residuals:                       87   BIC:                             904.2
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      -3.1111      3.657     -0.851      0.397     -10.379       4.157
CO_distance_from_max_scaled     1.2533      3.709      0.338      0.736      -6.119       8.626
average_distance_scaled         0.1017      3.709      0.027      0.978      -7.271       7.474
==============================================================================
Omnibus:                      140.150   Durbin-Watson:                   2.006
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                8.361
Skew:                           0.089   Prob(JB):                       0.0153
Kurtosis:                       1.517   Cond. No.                         1.18
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for AG, question 1:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_1   R-squared:                       0.001
Model:                             OLS   Adj. R-squared:                 -0.022
Method:                  Least Squares   F-statistic:                   0.06090
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.941
Time:                         07:59:45   Log-Likelihood:                -421.60
No. Observations:                   90   AIC:                             849.2
Df Residuals:                       87   BIC:                             856.7
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                       9.2667      2.808      3.300      0.001       3.685      14.848
AG_distance_from_max_scaled    -0.3437      2.867     -0.120      0.905      -6.041       5.354
average_distance_scaled         0.9894      2.867      0.345      0.731      -4.708       6.687
==============================================================================
Omnibus:                        5.821   Durbin-Watson:                   1.892
Prob(Omnibus):                  0.054   Jarque-Bera (JB):                5.091
Skew:                          -0.497   Prob(JB):                       0.0784
Kurtosis:                       2.394   Cond. No.                         1.23
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for AG, question 2:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_2   R-squared:                       0.011
Model:                             OLS   Adj. R-squared:                 -0.012
Method:                  Least Squares   F-statistic:                    0.4819
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.619
Time:                         07:59:45   Log-Likelihood:                -412.84
No. Observations:                   90   AIC:                             831.7
Df Residuals:                       87   BIC:                             839.2
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                      11.7111      2.548      4.597      0.000       6.647      16.775
AG_distance_from_max_scaled    -2.2699      2.601     -0.873      0.385      -7.439       2.899
average_distance_scaled        -0.6897      2.601     -0.265      0.791      -5.859       4.479
==============================================================================
Omnibus:                        3.126   Durbin-Watson:                   1.855
Prob(Omnibus):                  0.209   Jarque-Bera (JB):                3.118
Skew:                          -0.426   Prob(JB):                        0.210
Kurtosis:                       2.675   Cond. No.                         1.23
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for AG, question 3:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_3   R-squared:                       0.003
Model:                             OLS   Adj. R-squared:                 -0.020
Method:                  Least Squares   F-statistic:                    0.1328
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.876
Time:                         07:59:45   Log-Likelihood:                -426.72
No. Observations:                   90   AIC:                             859.4
Df Residuals:                       87   BIC:                             866.9
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                       9.4222      2.973      3.170      0.002       3.514      15.331
AG_distance_from_max_scaled     0.0561      3.035      0.019      0.985      -5.975       6.088
average_distance_scaled        -1.5423      3.035     -0.508      0.613      -7.574       4.489
==============================================================================
Omnibus:                        6.489   Durbin-Watson:                   1.944
Prob(Omnibus):                  0.039   Jarque-Bera (JB):                4.136
Skew:                          -0.358   Prob(JB):                        0.126
Kurtosis:                       2.231   Cond. No.                         1.23
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for AG, question 4:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_4   R-squared:                       0.001
Model:                             OLS   Adj. R-squared:                 -0.022
Method:                  Least Squares   F-statistic:                   0.02770
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.973
Time:                         07:59:45   Log-Likelihood:                -420.08
No. Observations:                   90   AIC:                             846.2
Df Residuals:                       87   BIC:                             853.7
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                       7.4111      2.761      2.684      0.009       1.923      12.899
AG_distance_from_max_scaled     0.6335      2.819      0.225      0.823      -4.969       6.236
average_distance_scaled         0.0657      2.819      0.023      0.981      -5.537       5.668
==============================================================================
Omnibus:                        3.105   Durbin-Watson:                   1.897
Prob(Omnibus):                  0.212   Jarque-Bera (JB):                2.529
Skew:                          -0.287   Prob(JB):                        0.282
Kurtosis:                       2.412   Cond. No.                         1.23
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for AG, question 5:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_5   R-squared:                       0.002
Model:                             OLS   Adj. R-squared:                 -0.021
Method:                  Least Squares   F-statistic:                    0.1017
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.903
Time:                         07:59:45   Log-Likelihood:                -423.28
No. Observations:                   90   AIC:                             852.6
Df Residuals:                       87   BIC:                             860.1
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                       7.9111      2.861      2.765      0.007       2.225      13.598
AG_distance_from_max_scaled     0.9945      2.920      0.341      0.734      -4.810       6.799
average_distance_scaled        -1.0460      2.920     -0.358      0.721      -6.851       4.759
==============================================================================
Omnibus:                        7.231   Durbin-Watson:                   1.966
Prob(Omnibus):                  0.027   Jarque-Bera (JB):                3.538
Skew:                          -0.237   Prob(JB):                        0.171
Kurtosis:                       2.152   Cond. No.                         1.23
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for AG, question 6:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_6   R-squared:                       0.002
Model:                             OLS   Adj. R-squared:                 -0.021
Method:                  Least Squares   F-statistic:                   0.08606
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.918
Time:                         07:59:45   Log-Likelihood:                -429.82
No. Observations:                   90   AIC:                             865.6
Df Residuals:                       87   BIC:                             873.1
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
Intercept                       9.5333      3.077      3.099      0.003       3.418      15.648
AG_distance_from_max_scaled     1.1107      3.141      0.354      0.724      -5.131       7.353
average_distance_scaled        -0.8901      3.141     -0.283      0.778      -7.132       5.352
==============================================================================
Omnibus:                        9.406   Durbin-Watson:                   1.786
Prob(Omnibus):                  0.009   Jarque-Bera (JB):                4.671
Skew:                          -0.337   Prob(JB):                       0.0968
Kurtosis:                       2.111   Cond. No.                         1.23
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

# 3. 多面的効果分析
effect_questions = {
    'preference': 1,
    'believability': 2,
    'overall_liking': 3,
    'persuasiveness': 4,
    'interest': 5,
    'click_likelihood': 6
}

def analyze_effect(effect):
    question_num = effect_questions[effect]
    for trait in ['EX', 'OP', 'CO', 'AG']:
        model = smf.ols(f'{trait}_relative_score_{question_num} ~ average_distance_scaled', data=data)
        results = model.fit()
        print(f"\nResults for {effect} ({trait}):")
        print(results.summary())

for effect in effect_questions.keys():
    analyze_effect(effect)

Results for preference (EX):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_1   R-squared:                       0.052
Model:                             OLS   Adj. R-squared:                  0.041
Method:                  Least Squares   F-statistic:                     4.847
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):             0.0303
Time:                         07:59:45   Log-Likelihood:                -440.57
No. Observations:                   90   AIC:                             885.1
Df Residuals:                       88   BIC:                             890.1
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                   3.3444      3.447      0.970      0.335      -3.506      10.195
average_distance_scaled     7.5893      3.447      2.202      0.030       0.739      14.440
==============================================================================
Omnibus:                       58.365   Durbin-Watson:                   1.767
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.043
Skew:                          -0.132   Prob(JB):                       0.0296
Kurtosis:                       1.655   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for preference (OP):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_1   R-squared:                       0.001
Model:                             OLS   Adj. R-squared:                 -0.011
Method:                  Least Squares   F-statistic:                   0.06089
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.806
Time:                         07:59:45   Log-Likelihood:                -438.93
No. Observations:                   90   AIC:                             881.9
Df Residuals:                       88   BIC:                             886.9
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                  -6.0556      3.385     -1.789      0.077     -12.783       0.672
average_distance_scaled     0.8353      3.385      0.247      0.806      -5.892       7.563
==============================================================================
Omnibus:                       53.483   Durbin-Watson:                   1.935
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.467
Skew:                           0.242   Prob(JB):                       0.0239
Kurtosis:                       1.674   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for preference (CO):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_1   R-squared:                       0.000
Model:                             OLS   Adj. R-squared:                 -0.011
Method:                  Least Squares   F-statistic:                 3.261e-05
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.995
Time:                         07:59:45   Log-Likelihood:                -436.44
No. Observations:                   90   AIC:                             876.9
Df Residuals:                       88   BIC:                             881.9
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                  -1.8444      3.293     -0.560      0.577      -8.388       4.699
average_distance_scaled    -0.0188      3.293     -0.006      0.995      -6.562       6.525
==============================================================================
Omnibus:                       46.111   Durbin-Watson:                   2.129
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                6.375
Skew:                           0.050   Prob(JB):                       0.0413
Kurtosis:                       1.700   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for preference (AG):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_1   R-squared:                       0.001
Model:                             OLS   Adj. R-squared:                 -0.010
Method:                  Least Squares   F-statistic:                    0.1086
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.742
Time:                         07:59:45   Log-Likelihood:                -421.61
No. Observations:                   90   AIC:                             847.2
Df Residuals:                       88   BIC:                             852.2
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                   9.2667      2.792      3.318      0.001       3.717      14.816
average_distance_scaled     0.9205      2.792      0.330      0.742      -4.629       6.470
==============================================================================
Omnibus:                        5.754   Durbin-Watson:                   1.890
Prob(Omnibus):                  0.056   Jarque-Bera (JB):                5.005
Skew:                          -0.491   Prob(JB):                       0.0819
Kurtosis:                       2.391   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for believability (EX):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_2   R-squared:                       0.043
Model:                             OLS   Adj. R-squared:                  0.032
Method:                  Least Squares   F-statistic:                     3.973
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):             0.0493
Time:                         07:59:45   Log-Likelihood:                -431.84
No. Observations:                   90   AIC:                             867.7
Df Residuals:                       88   BIC:                             872.7
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                   1.1111      3.129      0.355      0.723      -5.107       7.329
average_distance_scaled     6.2362      3.129      1.993      0.049       0.018      12.454
==============================================================================
Omnibus:                       17.112   Durbin-Watson:                   1.774
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                4.657
Skew:                          -0.143   Prob(JB):                       0.0974
Kurtosis:                       1.923   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for believability (OP):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_2   R-squared:                       0.000
Model:                             OLS   Adj. R-squared:                 -0.011
Method:                  Least Squares   F-statistic:                   0.02409
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.877
Time:                         07:59:45   Log-Likelihood:                -422.10
No. Observations:                   90   AIC:                             848.2
Df Residuals:                       88   BIC:                             853.2
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                  -4.2111      2.808     -1.500      0.137      -9.791       1.369
average_distance_scaled     0.4358      2.808      0.155      0.877      -5.144       6.016
==============================================================================
Omnibus:                        4.169   Durbin-Watson:                   2.258
Prob(Omnibus):                  0.124   Jarque-Bera (JB):                2.161
Skew:                           0.031   Prob(JB):                        0.339
Kurtosis:                       2.243   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for believability (CO):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_2   R-squared:                       0.004
Model:                             OLS   Adj. R-squared:                 -0.008
Method:                  Least Squares   F-statistic:                    0.3249
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.570
Time:                         07:59:45   Log-Likelihood:                -428.24
No. Observations:                   90   AIC:                             860.5
Df Residuals:                       88   BIC:                             865.5
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                  -0.8667      3.006     -0.288      0.774      -6.840       5.107
average_distance_scaled    -1.7134      3.006     -0.570      0.570      -7.687       4.260
==============================================================================
Omnibus:                        6.041   Durbin-Watson:                   2.063
Prob(Omnibus):                  0.049   Jarque-Bera (JB):                2.693
Skew:                           0.071   Prob(JB):                        0.260
Kurtosis:                       2.165   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for believability (AG):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_2   R-squared:                       0.002
Model:                             OLS   Adj. R-squared:                 -0.009
Method:                  Least Squares   F-statistic:                    0.2026
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.654
Time:                         07:59:45   Log-Likelihood:                -413.23
No. Observations:                   90   AIC:                             830.5
Df Residuals:                       88   BIC:                             835.5
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                  11.7111      2.544      4.603      0.000       6.655      16.767
average_distance_scaled    -1.1452      2.544     -0.450      0.654      -6.201       3.911
==============================================================================
Omnibus:                        2.828   Durbin-Watson:                   1.840
Prob(Omnibus):                  0.243   Jarque-Bera (JB):                2.795
Skew:                          -0.386   Prob(JB):                        0.247
Kurtosis:                       2.613   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for overall_liking (EX):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_3   R-squared:                       0.056
Model:                             OLS   Adj. R-squared:                  0.045
Method:                  Least Squares   F-statistic:                     5.205
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):             0.0249
Time:                         07:59:45   Log-Likelihood:                -442.67
No. Observations:                   90   AIC:                             889.3
Df Residuals:                       88   BIC:                             894.3
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                   2.7778      3.529      0.787      0.433      -4.235       9.790
average_distance_scaled     8.0503      3.529      2.281      0.025       1.038      15.063
==============================================================================
Omnibus:                      102.385   Durbin-Watson:                   1.812
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                8.055
Skew:                          -0.138   Prob(JB):                       0.0178
Kurtosis:                       1.561   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for overall_liking (OP):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_3   R-squared:                       0.000
Model:                             OLS   Adj. R-squared:                 -0.011
Method:                  Least Squares   F-statistic:                  0.004343
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.948
Time:                         07:59:45   Log-Likelihood:                -440.68
No. Observations:                   90   AIC:                             885.4
Df Residuals:                       88   BIC:                             890.4
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                  -6.9000      3.451     -1.999      0.049     -13.759      -0.041
average_distance_scaled    -0.2274      3.451     -0.066      0.948      -7.086       6.632
==============================================================================
Omnibus:                       53.076   Durbin-Watson:                   1.962
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.521
Skew:                           0.252   Prob(JB):                       0.0233
Kurtosis:                       1.676   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for overall_liking (CO):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_3   R-squared:                       0.002
Model:                             OLS   Adj. R-squared:                 -0.010
Method:                  Least Squares   F-statistic:                    0.1537
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.696
Time:                         07:59:45   Log-Likelihood:                -439.73
No. Observations:                   90   AIC:                             883.5
Df Residuals:                       88   BIC:                             888.5
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                  -3.6444      3.415     -1.067      0.289     -10.431       3.143
average_distance_scaled     1.3391      3.415      0.392      0.696      -5.448       8.126
==============================================================================
Omnibus:                       73.691   Durbin-Watson:                   1.995
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.410
Skew:                           0.115   Prob(JB):                       0.0246
Kurtosis:                       1.613   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for overall_liking (AG):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_3   R-squared:                       0.003
Model:                             OLS   Adj. R-squared:                 -0.008
Method:                  Least Squares   F-statistic:                    0.2683
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.606
Time:                         07:59:45   Log-Likelihood:                -426.73
No. Observations:                   90   AIC:                             857.5
Df Residuals:                       88   BIC:                             862.4
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                   9.4222      2.956      3.188      0.002       3.548      15.296
average_distance_scaled    -1.5311      2.956     -0.518      0.606      -7.405       4.343
==============================================================================
Omnibus:                        6.513   Durbin-Watson:                   1.945
Prob(Omnibus):                  0.039   Jarque-Bera (JB):                4.149
Skew:                          -0.359   Prob(JB):                        0.126
Kurtosis:                       2.231   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for persuasiveness (EX):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_4   R-squared:                       0.049
Model:                             OLS   Adj. R-squared:                  0.038
Method:                  Least Squares   F-statistic:                     4.559
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):             0.0355
Time:                         07:59:45   Log-Likelihood:                -437.43
No. Observations:                   90   AIC:                             878.9
Df Residuals:                       88   BIC:                             883.9
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                   0.6222      3.329      0.187      0.852      -5.994       7.238
average_distance_scaled     7.1080      3.329      2.135      0.036       0.492      13.724
==============================================================================
Omnibus:                       51.766   Durbin-Watson:                   1.705
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                6.638
Skew:                          -0.072   Prob(JB):                       0.0362
Kurtosis:                       1.677   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for persuasiveness (OP):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_4   R-squared:                       0.007
Model:                             OLS   Adj. R-squared:                 -0.005
Method:                  Least Squares   F-statistic:                    0.5797
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.448
Time:                         07:59:45   Log-Likelihood:                -432.98
No. Observations:                   90   AIC:                             870.0
Df Residuals:                       88   BIC:                             875.0
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                  -8.5222      3.168     -2.690      0.009     -14.819      -2.226
average_distance_scaled     2.4124      3.168      0.761      0.448      -3.884       8.709
==============================================================================
Omnibus:                       18.920   Durbin-Watson:                   2.008
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.873
Skew:                           0.312   Prob(JB):                       0.0531
Kurtosis:                       1.916   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for persuasiveness (CO):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_4   R-squared:                       0.003
Model:                             OLS   Adj. R-squared:                 -0.009
Method:                  Least Squares   F-statistic:                    0.2292
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.633
Time:                         07:59:45   Log-Likelihood:                -434.10
No. Observations:                   90   AIC:                             872.2
Df Residuals:                       88   BIC:                             877.2
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                  -2.8444      3.208     -0.887      0.378      -9.220       3.531
average_distance_scaled     1.5358      3.208      0.479      0.633      -4.840       7.911
==============================================================================
Omnibus:                       25.062   Durbin-Watson:                   2.018
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.233
Skew:                           0.086   Prob(JB):                       0.0731
Kurtosis:                       1.831   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for persuasiveness (AG):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_4   R-squared:                       0.000
Model:                             OLS   Adj. R-squared:                 -0.011
Method:                  Least Squares   F-statistic:                  0.004931
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.944
Time:                         07:59:45   Log-Likelihood:                -420.11
No. Observations:                   90   AIC:                             844.2
Df Residuals:                       88   BIC:                             849.2
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                   7.4111      2.746      2.699      0.008       1.953      12.869
average_distance_scaled     0.1929      2.746      0.070      0.944      -5.265       5.651
==============================================================================
Omnibus:                        3.341   Durbin-Watson:                   1.903
Prob(Omnibus):                  0.188   Jarque-Bera (JB):                2.687
Skew:                          -0.299   Prob(JB):                        0.261
Kurtosis:                       2.400   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for interest (EX):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_5   R-squared:                       0.073
Model:                             OLS   Adj. R-squared:                  0.062
Method:                  Least Squares   F-statistic:                     6.926
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):             0.0100
Time:                         07:59:45   Log-Likelihood:                -441.60
No. Observations:                   90   AIC:                             887.2
Df Residuals:                       88   BIC:                             892.2
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                   3.3222      3.487      0.953      0.343      -3.608      10.252
average_distance_scaled     9.1773      3.487      2.632      0.010       2.247      16.107
==============================================================================
Omnibus:                       65.588   Durbin-Watson:                   1.726
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.624
Skew:                          -0.206   Prob(JB):                       0.0221
Kurtosis:                       1.635   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for interest (OP):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_5   R-squared:                       0.002
Model:                             OLS   Adj. R-squared:                 -0.010
Method:                  Least Squares   F-statistic:                    0.1418
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.707
Time:                         07:59:45   Log-Likelihood:                -435.75
No. Observations:                   90   AIC:                             875.5
Df Residuals:                       88   BIC:                             880.5
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                  -8.2333      3.268     -2.520      0.014     -14.727      -1.739
average_distance_scaled     1.2304      3.268      0.377      0.707      -5.263       7.724
==============================================================================
Omnibus:                       18.819   Durbin-Watson:                   2.054
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                6.451
Skew:                           0.377   Prob(JB):                       0.0397
Kurtosis:                       1.927   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for interest (CO):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_5   R-squared:                       0.002
Model:                             OLS   Adj. R-squared:                 -0.010
Method:                  Least Squares   F-statistic:                    0.1616
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.689
Time:                         07:59:45   Log-Likelihood:                -438.04
No. Observations:                   90   AIC:                             880.1
Df Residuals:                       88   BIC:                             885.1
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                  -2.1556      3.352     -0.643      0.522      -8.816       4.505
average_distance_scaled     1.3473      3.352      0.402      0.689      -5.314       8.008
==============================================================================
Omnibus:                       33.450   Durbin-Watson:                   2.034
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.828
Skew:                           0.093   Prob(JB):                       0.0543
Kurtosis:                       1.767   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for interest (AG):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_5   R-squared:                       0.001
Model:                             OLS   Adj. R-squared:                 -0.010
Method:                  Least Squares   F-statistic:                   0.08840
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.767
Time:                         07:59:45   Log-Likelihood:                -423.34
No. Observations:                   90   AIC:                             850.7
Df Residuals:                       88   BIC:                             855.7
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                   7.9111      2.847      2.779      0.007       2.254      13.568
average_distance_scaled    -0.8464      2.847     -0.297      0.767      -6.503       4.811
==============================================================================
Omnibus:                        7.366   Durbin-Watson:                   1.975
Prob(Omnibus):                  0.025   Jarque-Bera (JB):                3.727
Skew:                          -0.265   Prob(JB):                        0.155
Kurtosis:                       2.156   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for click_likelihood (EX):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_6   R-squared:                       0.044
Model:                             OLS   Adj. R-squared:                  0.033
Method:                  Least Squares   F-statistic:                     4.074
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):             0.0466
Time:                         07:59:45   Log-Likelihood:                -445.61
No. Observations:                   90   AIC:                             895.2
Df Residuals:                       88   BIC:                             900.2
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                   3.0333      3.646      0.832      0.408      -4.212      10.279
average_distance_scaled     7.3593      3.646      2.018      0.047       0.114      14.605
==============================================================================
Omnibus:                       82.639   Durbin-Watson:                   1.770
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.971
Skew:                          -0.194   Prob(JB):                       0.0186
Kurtosis:                       1.595   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for click_likelihood (OP):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_6   R-squared:                       0.003
Model:                             OLS   Adj. R-squared:                 -0.009
Method:                  Least Squares   F-statistic:                    0.2370
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.628
Time:                         07:59:45   Log-Likelihood:                -440.04
No. Observations:                   90   AIC:                             884.1
Df Residuals:                       88   BIC:                             889.1
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                  -8.0444      3.427     -2.347      0.021     -14.855      -1.234
average_distance_scaled     1.6685      3.427      0.487      0.628      -5.142       8.479
==============================================================================
Omnibus:                       50.413   Durbin-Watson:                   1.909
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.635
Skew:                           0.280   Prob(JB):                       0.0220
Kurtosis:                       1.687   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for click_likelihood (CO):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_6   R-squared:                       0.000
Model:                             OLS   Adj. R-squared:                 -0.011
Method:                  Least Squares   F-statistic:                  0.007353
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.932
Time:                         07:59:45   Log-Likelihood:                -445.42
No. Observations:                   90   AIC:                             894.8
Df Residuals:                       88   BIC:                             899.8
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                  -3.1111      3.638     -0.855      0.395     -10.341       4.119
average_distance_scaled     0.3120      3.638      0.086      0.932      -6.918       7.542
==============================================================================
Omnibus:                      158.113   Durbin-Watson:                   2.020
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                8.490
Skew:                           0.074   Prob(JB):                       0.0143
Kurtosis:                       1.503   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Results for click_likelihood (AG):
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_6   R-squared:                       0.001
Model:                             OLS   Adj. R-squared:                 -0.011
Method:                  Least Squares   F-statistic:                   0.04750
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.828
Time:                         07:59:45   Log-Likelihood:                -429.88
No. Observations:                   90   AIC:                             863.8
Df Residuals:                       88   BIC:                             868.8
Df Model:                            1                                         
Covariance Type:             nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                   9.5333      3.061      3.114      0.002       3.450      15.617
average_distance_scaled    -0.6672      3.061     -0.218      0.828      -6.751       5.416
==============================================================================
Omnibus:                        9.644   Durbin-Watson:                   1.794
Prob(Omnibus):                  0.008   Jarque-Bera (JB):                4.922
Skew:                          -0.362   Prob(JB):                       0.0854
Kurtosis:                       2.112   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

# 4. 距離の影響分析
for trait in ['EX', 'OP', 'CO', 'AG']:
    data[f'{trait}_distance_diff'] = data[f'{trait}_distance_from_max'] - data[f'{trait}_distance']

def analyze_distances(trait):
    model = smf.ols(f'{trait}_relative_score_1 ~ {trait}_distance + {trait}_distance_from_max + '
                    f'{trait}_distance_diff', data=data)
    results = model.fit()
    print(f"\nDistance analysis for {trait}:")
    print(results.summary())

for trait in ['EX', 'OP', 'CO', 'AG']:
    analyze_distances(trait)

Distance analysis for EX:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_1   R-squared:                       0.107
Model:                             OLS   Adj. R-squared:                  0.087
Method:                  Least Squares   F-statistic:                     5.235
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):            0.00713
Time:                         07:59:45   Log-Likelihood:                -437.86
No. Observations:                   90   AIC:                             881.7
Df Residuals:                       87   BIC:                             889.2
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept              -21.9229      8.821     -2.485      0.015     -39.456      -4.390
EX_distance             11.9212      4.446      2.682      0.009       3.085      20.757
EX_distance_from_max    10.0446      3.121      3.218      0.002       3.841      16.248
EX_distance_diff        -1.8766      2.873     -0.653      0.515      -7.588       3.834
==============================================================================
Omnibus:                       10.074   Durbin-Watson:                   1.802
Prob(Omnibus):                  0.006   Jarque-Bera (JB):                3.523
Skew:                          -0.086   Prob(JB):                        0.172
Kurtosis:                       2.046   Cond. No.                     1.94e+16
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
[2] The smallest eigenvalue is 1.85e-30. This might indicate that there are
strong multicollinearity problems or that the design matrix is singular.

Distance analysis for OP:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_1   R-squared:                       0.022
Model:                             OLS   Adj. R-squared:                 -0.001
Method:                  Least Squares   F-statistic:                    0.9588
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.387
Time:                         07:59:45   Log-Likelihood:                -437.98
No. Observations:                   90   AIC:                             882.0
Df Residuals:                       87   BIC:                             889.5
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept              -14.0934      7.130     -1.977      0.051     -28.266       0.079
OP_distance              2.3980      5.134      0.467      0.642      -7.807      12.603
OP_distance_from_max     4.3457      3.451      1.259      0.211      -2.514      11.205
OP_distance_diff         1.9477      3.677      0.530      0.598      -5.361       9.257
==============================================================================
Omnibus:                       37.534   Durbin-Watson:                   1.885
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                6.940
Skew:                           0.267   Prob(JB):                       0.0311
Kurtosis:                       1.749   Cond. No.                     1.09e+16
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
[2] The smallest eigenvalue is 2.96e-30. This might indicate that there are
strong multicollinearity problems or that the design matrix is singular.

Distance analysis for CO:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_1   R-squared:                       0.012
Model:                             OLS   Adj. R-squared:                 -0.010
Method:                  Least Squares   F-statistic:                    0.5435
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.583
Time:                         07:59:46   Log-Likelihood:                -435.88
No. Observations:                   90   AIC:                             877.8
Df Residuals:                       87   BIC:                             885.3
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept               -7.2744      6.646     -1.094      0.277     -20.485       5.936
CO_distance              5.3361      5.215      1.023      0.309      -5.029      15.701
CO_distance_from_max     3.0448      3.442      0.885      0.379      -3.796       9.885
CO_distance_diff        -2.2913      3.577     -0.641      0.523      -9.401       4.818
==============================================================================
Omnibus:                       41.883   Durbin-Watson:                   2.111
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                6.217
Skew:                           0.068   Prob(JB):                       0.0447
Kurtosis:                       1.720   Cond. No.                     1.09e+16
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
[2] The smallest eigenvalue is 2.43e-30. This might indicate that there are
strong multicollinearity problems or that the design matrix is singular.

Distance analysis for AG:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_1   R-squared:                       0.048
Model:                             OLS   Adj. R-squared:                  0.026
Method:                  Least Squares   F-statistic:                     2.202
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.117
Time:                         07:59:46   Log-Likelihood:                -419.44
No. Observations:                   90   AIC:                             844.9
Df Residuals:                       87   BIC:                             852.4
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept                3.4629      5.814      0.596      0.553      -8.092      15.018
AG_distance              9.3894      4.695      2.000      0.049       0.058      18.721
AG_distance_from_max     3.1495      3.050      1.032      0.305      -2.914       9.213
AG_distance_diff        -6.2400      3.220     -1.938      0.056     -12.639       0.159
==============================================================================
Omnibus:                        7.125   Durbin-Watson:                   2.010
Prob(Omnibus):                  0.028   Jarque-Bera (JB):                7.317
Skew:                          -0.697   Prob(JB):                       0.0258
Kurtosis:                       2.910   Cond. No.                     1.58e+16
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
[2] The smallest eigenvalue is 1.18e-30. This might indicate that there are
strong multicollinearity problems or that the design matrix is singular.

Distance analysis for AG:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_1   R-squared:                       0.048
Model:                             OLS   Adj. R-squared:                  0.026
Method:                  Least Squares   F-statistic:                     2.202
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.117
Time:                         07:59:46   Log-Likelihood:                -419.44
No. Observations:                   90   AIC:                             844.9
Df Residuals:                       87   BIC:                             852.4
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept                3.4629      5.814      0.596      0.553      -8.092      15.018
AG_distance              9.3894      4.695      2.000      0.049       0.058      18.721
AG_distance_from_max     3.1495      3.050      1.032      0.305      -2.914       9.213
AG_distance_diff        -6.2400      3.220     -1.938      0.056     -12.639       0.159
==============================================================================
Omnibus:                        7.125   Durbin-Watson:                   2.010
Prob(Omnibus):                  0.028   Jarque-Bera (JB):                7.317
Skew:                          -0.697   Prob(JB):                       0.0258
Kurtosis:                       2.910   Cond. No.                     1.58e+16
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
[2] The smallest eigenvalue is 1.18e-30. This might indicate that there are
strong multicollinearity problems or that the design matrix is singular.

# 5. セグメント分析
def create_segments(trait):
    kmeans = KMeans(n_clusters=3, random_state=42)
    data[f'{trait}_segment'] = kmeans.fit_predict(data[[f'{trait}_distance_from_max_scaled', 'average_distance_scaled']])

for trait in ['EX', 'OP', 'CO', 'AG']:
    create_segments(trait)

def analyze_segments(trait):
    model = smf.ols(f'{trait}_relative_score_1 ~ C({trait}_segment)', data=data)
    results = model.fit()
    print(f"\nSegment analysis for {trait}:")
    print(results.summary())

for trait in ['EX', 'OP', 'CO', 'AG']:
    analyze_segments(trait)

Segment analysis for EX:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_1   R-squared:                       0.044
Model:                             OLS   Adj. R-squared:                  0.022
Method:                  Least Squares   F-statistic:                     1.980
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.144
Time:                         07:59:50   Log-Likelihood:                -440.97
No. Observations:                   90   AIC:                             887.9
Df Residuals:                       87   BIC:                             895.4
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
======================================================================================
                         coef    std err          t      P>|t|      [0.025      0.975]
--------------------------------------------------------------------------------------
Intercept             -0.4706      5.666     -0.083      0.934     -11.733      10.792
C(EX_segment)[T.1]     0.4436      7.849      0.057      0.955     -15.158      16.045
C(EX_segment)[T.2]    17.2074      9.464      1.818      0.072      -1.603      36.018
==============================================================================
Omnibus:                       58.890   Durbin-Watson:                   1.737
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.007
Skew:                          -0.118   Prob(JB):                       0.0301
Kurtosis:                       1.653   Cond. No.                         3.67
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Segment analysis for OP:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_1   R-squared:                       0.032
Model:                             OLS   Adj. R-squared:                  0.010
Method:                  Least Squares   F-statistic:                     1.427
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.246
Time:                         07:59:50   Log-Likelihood:                -437.51
No. Observations:                   90   AIC:                             881.0
Df Residuals:                       87   BIC:                             888.5
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
======================================================================================
                         coef    std err          t      P>|t|      [0.025      0.975]
--------------------------------------------------------------------------------------
Intercept              0.6333      5.804      0.109      0.913     -10.904      12.170
C(OP_segment)[T.1]   -16.6333     10.054     -1.654      0.102     -36.616       3.349
C(OP_segment)[T.2]    -7.8333      7.494     -1.045      0.299     -22.728       7.061
==============================================================================
Omnibus:                       38.891   Durbin-Watson:                   1.913
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.080
Skew:                           0.276   Prob(JB):                       0.0290
Kurtosis:                       1.741   Cond. No.                         4.05
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Segment analysis for CO:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_1   R-squared:                       0.012
Model:                             OLS   Adj. R-squared:                 -0.011
Method:                  Least Squares   F-statistic:                    0.5109
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.602
Time:                         07:59:50   Log-Likelihood:                -435.91
No. Observations:                   90   AIC:                             877.8
Df Residuals:                       87   BIC:                             885.3
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
======================================================================================
                         coef    std err          t      P>|t|      [0.025      0.975]
--------------------------------------------------------------------------------------
Intercept             -0.1429      4.819     -0.030      0.976      -9.722       9.436
C(CO_segment)[T.1]    -6.2905      7.466     -0.843      0.402     -21.130       8.549
C(CO_segment)[T.2]     1.9762      8.799      0.225      0.823     -15.512      19.465
==============================================================================
Omnibus:                       33.743   Durbin-Watson:                   2.117
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.828
Skew:                           0.086   Prob(JB):                       0.0543
Kurtosis:                       1.765   Cond. No.                         3.37
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Segment analysis for AG:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_1   R-squared:                       0.004
Model:                             OLS   Adj. R-squared:                 -0.019
Method:                  Least Squares   F-statistic:                    0.1890
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.828
Time:                         07:59:50   Log-Likelihood:                -421.47
No. Observations:                   90   AIC:                             848.9
Df Residuals:                       87   BIC:                             856.4
Df Model:                            2                                         
Covariance Type:             nonrobust                                         
======================================================================================
                         coef    std err          t      P>|t|      [0.025      0.975]
--------------------------------------------------------------------------------------
Intercept              6.4545      5.672      1.138      0.258      -4.818      17.727
C(AG_segment)[T.1]     4.5799      7.521      0.609      0.544     -10.369      19.529
C(AG_segment)[T.2]     3.0839      7.093      0.435      0.665     -11.015      17.182
==============================================================================
Omnibus:                        5.899   Durbin-Watson:                   1.880
Prob(Omnibus):                  0.052   Jarque-Bera (JB):                5.173
Skew:                          -0.503   Prob(JB):                       0.0753
Kurtosis:                       2.394   Cond. No.                         4.33
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

# 6. 交互作用効果の分析
def analyze_interactions(trait):
    model = smf.ols(f'{trait}_relative_score_1 ~ {trait}_distance_from_max_scaled * average_distance_scaled', data=data)
    results = model.fit()
    print(f"\nInteraction analysis for {trait}:")
    print(results.summary())

for trait in ['EX', 'OP', 'CO', 'AG']:
    analyze_interactions(trait)

# 可視化
plt.figure(figsize=(12, 8))
for i, trait in enumerate(['EX', 'OP', 'CO', 'AG']):
    plt.subplot(2, 2, i+1)
    sns.scatterplot(x=f'{trait}_distance_from_max_scaled', y=f'{trait}_relative_score_1', 
                    hue='average_distance_scaled', data=data)
    plt.title(f'{trait} Analysis')
plt.tight_layout()
plt.show()

# 可視化2
plt.figure(figsize=(12, 8))

# 全てのTraitの中での最大値と最小値を計算
x_min = min(data[f'{trait}_distance_from_max'].min() for trait in ['EX', 'OP', 'CO', 'AG'])
x_max = max(data[f'{trait}_distance_from_max'].max() for trait in ['EX', 'OP', 'CO', 'AG'])
y_min = data['average_distance'].min()
y_max = data['average_distance'].max()

# マージンを追加
margin = 0.1
x_range = x_max - x_min
y_range = y_max - y_min
x_min -= margin * x_range
x_max += margin * x_range
y_min -= margin * y_range
y_max += margin * y_range

for i, trait in enumerate(['EX', 'OP', 'CO', 'AG']):
    plt.subplot(2, 2, i+1)
    
    scatter = plt.scatter(data[f'{trait}_distance_from_max'], 
                          data['average_distance'],
                          c=data[f'{trait}_relative_score_1'], 
                          cmap='coolwarm', 
                          alpha=0.7)
    plt.colorbar(scatter, label=f'Perceived_personalization_score')
    plt.xlabel(f'{trait}_distance_from_max')
    plt.ylabel('average_distance')
    plt.title(f'{trait} Perceived Personalization Score Distribution')
    
    # 全てのチャートで同じX軸とY軸の範囲を設定
    plt.xlim(x_min, x_max)
    plt.ylim(y_min, y_max)

plt.tight_layout()
plt.show()

# 可視化3
plt.figure(figsize=(12, 8))

# 全てのTraitの中での最大値と最小値を計算
x_min = min(data[f'{trait}_distance_from_max_scaled'].min() for trait in ['EX', 'OP', 'CO', 'AG'])
x_max = max(data[f'{trait}_distance_from_max_scaled'].max() for trait in ['EX', 'OP', 'CO', 'AG'])
y_min = data['average_distance_scaled'].min()
y_max = data['average_distance_scaled'].max()

# マージンを追加
margin = 0.1
x_range = x_max - x_min
y_range = y_max - y_min
x_min -= margin * x_range
x_max += margin * x_range
y_min -= margin * y_range
y_max += margin * y_range

for i, trait in enumerate(['EX', 'OP', 'CO', 'AG']):
    plt.subplot(2, 2, i+1)
    
    scatter = plt.scatter(data[f'{trait}_distance_from_max_scaled'], 
                          data['average_distance_scaled'],
                          c=data[f'{trait}_relative_score_1'], 
                          cmap='coolwarm', 
                          alpha=0.7)
    plt.colorbar(scatter, label=f'Perceived_personalization_score')
    plt.xlabel(f'{trait}_distance_from_max_scaled')
    plt.ylabel('average_distance_scaled')
    plt.title(f'{trait} Perceived Personalization Score Distribution')
    
    # 全てのチャートで同じX軸とY軸の範囲を設定
    plt.xlim(x_min, x_max)
    plt.ylim(y_min, y_max)

plt.tight_layout()
plt.show()

Interaction analysis for EX:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_1   R-squared:                       0.108
Model:                             OLS   Adj. R-squared:                  0.077
Method:                  Least Squares   F-statistic:                     3.460
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):             0.0198
Time:                         08:59:01   Log-Likelihood:                -437.85
No. Observations:                   90   AIC:                             883.7
Df Residuals:                       86   BIC:                             893.7
Df Model:                            3                                         
Covariance Type:             nonrobust                                         
=======================================================================================================================
                                                          coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------------------------------
Intercept                                               3.7187      3.427      1.085      0.281      -3.095      10.532
EX_distance_from_max_scaled                             9.2367      4.213      2.193      0.031       0.862      17.611
average_distance_scaled                                 6.9733      3.486      2.001      0.049       0.044      13.902
EX_distance_from_max_scaled:average_distance_scaled    -2.6141      3.819     -0.685      0.495     -10.206       4.978
==============================================================================
Omnibus:                       18.410   Durbin-Watson:                   1.754
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                4.741
Skew:                          -0.127   Prob(JB):                       0.0934
Kurtosis:                       1.905   Cond. No.                         2.11
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Interaction analysis for OP:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_1   R-squared:                       0.070
Model:                             OLS   Adj. R-squared:                  0.038
Method:                  Least Squares   F-statistic:                     2.162
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):             0.0984
Time:                         08:59:01   Log-Likelihood:                -435.69
No. Observations:                   90   AIC:                             879.4
Df Residuals:                       86   BIC:                             889.4
Df Model:                            3                                         
Covariance Type:             nonrobust                                         
=======================================================================================================================
                                                          coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------------------------------
Intercept                                              -7.6060      3.383     -2.248      0.027     -14.332      -0.880
OP_distance_from_max_scaled                            -0.6529      4.214     -0.155      0.877      -9.030       7.724
average_distance_scaled                                 0.8907      3.412      0.261      0.795      -5.891       7.673
OP_distance_from_max_scaled:average_distance_scaled     7.4655      3.522      2.119      0.037       0.463      14.468
==============================================================================
Omnibus:                       42.283   Durbin-Watson:                   1.871
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                6.957
Skew:                           0.235   Prob(JB):                       0.0308
Kurtosis:                       1.722   Cond. No.                         2.17
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Interaction analysis for CO:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_1   R-squared:                       0.007
Model:                             OLS   Adj. R-squared:                 -0.027
Method:                  Least Squares   F-statistic:                    0.2161
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.885
Time:                         08:59:01   Log-Likelihood:                -436.10
No. Observations:                   90   AIC:                             880.2
Df Residuals:                       86   BIC:                             890.2
Df Model:                            3                                         
Covariance Type:             nonrobust                                         
=======================================================================================================================
                                                          coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------------------------------
Intercept                                              -1.4752      3.368     -0.438      0.662      -8.170       5.220
CO_distance_from_max_scaled                             3.0570      4.028      0.759      0.450      -4.949      11.064
average_distance_scaled                                 0.0336      3.404      0.010      0.992      -6.734       6.801
CO_distance_from_max_scaled:average_distance_scaled    -2.2010      3.432     -0.641      0.523      -9.024       4.622
==============================================================================
Omnibus:                       32.707   Durbin-Watson:                   2.139
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.705
Skew:                           0.056   Prob(JB):                       0.0577
Kurtosis:                       1.772   Cond. No.                         2.03
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

Interaction analysis for AG:
                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_1   R-squared:                       0.002
Model:                             OLS   Adj. R-squared:                 -0.033
Method:                  Least Squares   F-statistic:                   0.04428
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.988
Time:                         08:59:01   Log-Likelihood:                -421.59
No. Observations:                   90   AIC:                             851.2
Df Residuals:                       86   BIC:                             861.2
Df Model:                            3                                         
Covariance Type:             nonrobust                                         
=======================================================================================================================
                                                          coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------------------------------
Intercept                                               9.2085      2.872      3.206      0.002       3.499      14.918
AG_distance_from_max_scaled                            -0.4963      3.192     -0.156      0.877      -6.841       5.848
average_distance_scaled                                 0.9301      2.932      0.317      0.752      -4.898       6.758
AG_distance_from_max_scaled:average_distance_scaled     0.2900      2.602      0.111      0.912      -4.883       5.463
==============================================================================
Omnibus:                        5.765   Durbin-Watson:                   1.893
Prob(Omnibus):                  0.056   Jarque-Bera (JB):                5.048
Skew:                          -0.495   Prob(JB):                       0.0801
Kurtosis:                       2.396   Cond. No.                         1.83
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from scipy import stats
import statsmodels.api as sm
from statsmodels.stats.multitest import multipletests

def analyze_trait(trait, data):
    # 3x3のグリッドを作成
    x_percentiles = np.percentile(data[f'{trait}_distance_from_max_scaled'], [33, 67])
    y_percentiles = np.percentile(data['average_distance_scaled'], [33, 67])
    
    # プロット
    plt.figure(figsize=(10, 8))
    scatter = plt.scatter(data[f'{trait}_distance_from_max_scaled'], 
                          data['average_distance_scaled'],
                          c=data[f'{trait}_relative_score_1'], 
                          cmap='coolwarm', 
                          alpha=0.7)
    plt.colorbar(scatter, label='Perceived personalization score')
    plt.xlabel(f'{trait}_distance_from_max_scaled')
    plt.ylabel('average_distance_scaled')
    plt.title(f'{trait} Perceived Personalization Score Distribution')
    
    # グリッドラインを追加
    for x in x_percentiles:
        plt.axvline(x, color='grey', linestyle='--', alpha=0.5)
    for y in y_percentiles:
        plt.axhline(y, color='grey', linestyle='--', alpha=0.5)
    
    plt.tight_layout()
    plt.show()
    
    # セクションの定義
    def get_section(x, y):
        x_section = np.digitize(x, x_percentiles)
        y_section = np.digitize(y, y_percentiles)
        return y_section * 3 + x_section + 1
    
    data['section'] = get_section(data[f'{trait}_distance_from_max_scaled'], 
                                  data['average_distance_scaled'])
    
    # 回帰分析
    X = sm.add_constant(data[[f'{trait}_distance_from_max_scaled', 'average_distance_scaled']])
    X['interaction'] = data[f'{trait}_distance_from_max_scaled'] * data['average_distance_scaled']
    y = data[f'{trait}_relative_score_1']
    model = sm.OLS(y, X).fit()
    print(model.summary())
    
    # セクション間の比較
    sections_to_compare = [1, 3, 7, 9]  # 四隅のセクション
    section_scores = [data[data['section'] == s][f'{trait}_relative_score_1'] for s in sections_to_compare]
    
    # ANOVA
    f_statistic, p_value = stats.f_oneway(*section_scores)
    print(f"ANOVA results: F-statistic = {f_statistic}, p-value = {p_value}")
    
    # ポストホック分析（Tukey's HSD）
    from statsmodels.stats.multicomp import pairwise_tukeyhsd
    scores_flat = np.concatenate(section_scores)
    groups_flat = np.concatenate([[s] * len(scores) for s, scores in zip(sections_to_compare, section_scores)])
    tukey_results = pairwise_tukeyhsd(scores_flat, groups_flat)
    print(tukey_results)
    
    # 多重比較の補正
    _, corrected_p_values, _, _ = multipletests([pair[3] for pair in tukey_results._results_table.data[1:]], 
                                                method='fdr_bh')
    print("Corrected p-values:", corrected_p_values)

# 各特性に対して分析を実行
for trait in ['EX', 'OP', 'CO', 'AG']:
    analyze_trait(trait, data)

                             OLS Regression Results                            
===============================================================================
Dep. Variable:     EX_relative_score_1   R-squared:                       0.108
Model:                             OLS   Adj. R-squared:                  0.077
Method:                  Least Squares   F-statistic:                     3.460
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):             0.0198
Time:                         09:15:38   Log-Likelihood:                -437.85
No. Observations:                   90   AIC:                             883.7
Df Residuals:                       86   BIC:                             893.7
Df Model:                            3                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
const                           3.7187      3.427      1.085      0.281      -3.095      10.532
EX_distance_from_max_scaled     9.2367      4.213      2.193      0.031       0.862      17.611
average_distance_scaled         6.9733      3.486      2.001      0.049       0.044      13.902
interaction                    -2.6141      3.819     -0.685      0.495     -10.206       4.978
==============================================================================
Omnibus:                       18.410   Durbin-Watson:                   1.754
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                4.741
Skew:                          -0.127   Prob(JB):                       0.0934
Kurtosis:                       1.905   Cond. No.                         2.11
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
ANOVA results: F-statistic = 1.0782167125833009, p-value = 0.3713577803810246
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
     1      3  27.2222  0.564 -28.9627 83.4071  False
     1      7   9.6333 0.9654 -45.8448 65.1115  False
     1      9  25.0833 0.5831   -27.94 78.1067  False
     3      7 -17.5889 0.6145 -56.3117 21.1339  False
     3      9  -2.1389 0.9984 -37.2545 32.9767  False
     7      9    15.45 0.6136 -18.5233 49.4233  False
-----------------------------------------------------
Corrected p-values: [0.92175 0.9984  0.92175 0.92175 0.9984  0.92175]

                             OLS Regression Results                            
===============================================================================
Dep. Variable:     OP_relative_score_1   R-squared:                       0.070
Model:                             OLS   Adj. R-squared:                  0.038
Method:                  Least Squares   F-statistic:                     2.162
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):             0.0984
Time:                         09:15:39   Log-Likelihood:                -435.69
No. Observations:                   90   AIC:                             879.4
Df Residuals:                       86   BIC:                             889.4
Df Model:                            3                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
const                          -7.6060      3.383     -2.248      0.027     -14.332      -0.880
OP_distance_from_max_scaled    -0.6529      4.214     -0.155      0.877      -9.030       7.724
average_distance_scaled         0.8907      3.412      0.261      0.795      -5.891       7.673
interaction                     7.4655      3.522      2.119      0.037       0.463      14.468
==============================================================================
Omnibus:                       42.283   Durbin-Watson:                   1.871
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                6.957
Skew:                           0.235   Prob(JB):                       0.0308
Kurtosis:                       1.722   Cond. No.                         2.17
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
ANOVA results: F-statistic = 2.7626747055907366, p-value = 0.058031488320304726
  Multiple Comparison of Means - Tukey HSD, FWER=0.05  
=======================================================
group1 group2 meandiff p-adj    lower    upper   reject
-------------------------------------------------------
     1      3 -41.3571 0.4103 -112.7952  30.0809  False
     1      7 -23.0571 0.4949  -66.9655  20.8513  False
     1      9   8.0252 0.9476  -31.9881  48.0385  False
     3      7     18.3  0.889  -50.7157  87.3157  False
     3      9  49.3824  0.206  -17.2231 115.9878  False
     7      9  31.0824 0.1033   -4.4259  66.5906  False
-------------------------------------------------------
Corrected p-values: [0.74235 0.74235 0.9476  0.9476  0.618   0.618  ]

                             OLS Regression Results                            
===============================================================================
Dep. Variable:     CO_relative_score_1   R-squared:                       0.007
Model:                             OLS   Adj. R-squared:                 -0.027
Method:                  Least Squares   F-statistic:                    0.2161
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.885
Time:                         09:15:40   Log-Likelihood:                -436.10
No. Observations:                   90   AIC:                             880.2
Df Residuals:                       86   BIC:                             890.2
Df Model:                            3                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
const                          -1.4752      3.368     -0.438      0.662      -8.170       5.220
CO_distance_from_max_scaled     3.0570      4.028      0.759      0.450      -4.949      11.064
average_distance_scaled         0.0336      3.404      0.010      0.992      -6.734       6.801
interaction                    -2.2010      3.432     -0.641      0.523      -9.024       4.622
==============================================================================
Omnibus:                       32.707   Durbin-Watson:                   2.139
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.705
Skew:                           0.056   Prob(JB):                       0.0577
Kurtosis:                       1.772   Cond. No.                         2.03
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
ANOVA results: F-statistic = 0.6398566094497207, p-value = 0.5946230850706252
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
     1      3   25.125 0.6127 -30.0523 80.3023  False
     1      7  23.9773  0.612 -28.6322 76.5867  False
     1      9  24.3167 0.5722 -26.3878 75.0211  False
     3      7  -1.1477 0.9999 -43.0155 40.7201  False
     3      9  -0.8083 0.9999 -40.2557  38.639  False
     7      9   0.3394    1.0 -35.4282  36.107  False
-----------------------------------------------------
Corrected p-values: [1. 1. 1. 1. 1. 1.]

                             OLS Regression Results                            
===============================================================================
Dep. Variable:     AG_relative_score_1   R-squared:                       0.002
Model:                             OLS   Adj. R-squared:                 -0.033
Method:                  Least Squares   F-statistic:                   0.04428
Date:                 Thu, 25 Jul 2024   Prob (F-statistic):              0.988
Time:                         09:15:40   Log-Likelihood:                -421.59
No. Observations:                   90   AIC:                             851.2
Df Residuals:                       86   BIC:                             861.2
Df Model:                            3                                         
Covariance Type:             nonrobust                                         
===============================================================================================
                                  coef    std err          t      P>|t|      [0.025      0.975]
-----------------------------------------------------------------------------------------------
const                           9.2085      2.872      3.206      0.002       3.499      14.918
AG_distance_from_max_scaled    -0.4963      3.192     -0.156      0.877      -6.841       5.848
average_distance_scaled         0.9301      2.932      0.317      0.752      -4.898       6.758
interaction                     0.2900      2.602      0.111      0.912      -4.883       5.463
==============================================================================
Omnibus:                        5.765   Durbin-Watson:                   1.893
Prob(Omnibus):                  0.056   Jarque-Bera (JB):                5.048
Skew:                          -0.495   Prob(JB):                       0.0801
Kurtosis:                       2.396   Cond. No.                         1.83
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
ANOVA results: F-statistic = 0.35178865412898275, p-value = 0.7881234128801623
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
     1      3      5.6 0.9918 -47.8957 59.0957  False
     1      7  -2.4286 0.9992 -52.4124 47.5553  False
     1      9  -8.4211 0.9534 -52.2912 35.4491  False
     3      7  -8.0286 0.9658 -54.7234 38.6663  False
     3      9 -14.0211 0.7786 -54.1037 26.0616  False
     7      9  -5.9925 0.9669 -41.2517 29.2668  False
-----------------------------------------------------
Corrected p-values: [0.9992 0.9992 0.9992 0.9992 0.9992 0.9992]
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
     1      3      5.6 0.9918 -47.8957 59.0957  False
     1      7  -2.4286 0.9992 -52.4124 47.5553  False
     1      9  -8.4211 0.9534 -52.2912 35.4491  False
     3      7  -8.0286 0.9658 -54.7234 38.6663  False
     3      9 -14.0211 0.7786 -54.1037 26.0616  False
     7      9  -5.9925 0.9669 -41.2517 29.2668  False
-----------------------------------------------------
Corrected p-values: [0.9992 0.9992 0.9992 0.9992 0.9992 0.9992]

# 総合的な効果の分析
data['overall_blended_preference'] = data[[f'{trait}_relative_score_1' for trait in ['EX', 'OP', 'CO', 'AG']]].mean(axis=1)
model = smf.ols('overall_blended_preference ~ average_distance_scaled', data=data)
results = model.fit()
print("\nOverall Blended Preference Analysis:")
print(results.summary())

Overall Blended Preference Analysis:
                                OLS Regression Results                                
======================================================================================
Dep. Variable:     overall_blended_preference   R-squared:                       0.021
Model:                                    OLS   Adj. R-squared:                  0.010
Method:                         Least Squares   F-statistic:                     1.871
Date:                        Thu, 25 Jul 2024   Prob (F-statistic):              0.175
Time:                                07:59:52   Log-Likelihood:                -377.19
No. Observations:                          90   AIC:                             758.4
Df Residuals:                              88   BIC:                             763.4
Df Model:                                   1                                         
Covariance Type:                    nonrobust                                         
===========================================================================================
                              coef    std err          t      P>|t|      [0.025      0.975]
-------------------------------------------------------------------------------------------
Intercept                   1.1778      1.705      0.691      0.491      -2.210       4.566
average_distance_scaled     2.3316      1.705      1.368      0.175      -1.056       5.719
==============================================================================
Omnibus:                        0.996   Durbin-Watson:                   1.927
Prob(Omnibus):                  0.608   Jarque-Bera (JB):                1.068
Skew:                           0.235   Prob(JB):                        0.586
Kurtosis:                       2.747   Cond. No.                         1.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

sandbox 3 hlm

import pandas as pd
import numpy as np
import statsmodels.api as sm
import statsmodels.formula.api as smf
from scipy import stats
import matplotlib.pyplot as plt
import seaborn as sns

# データの読み込み（既存のデータフレームを 'data' とします）
data = merged_df

# 1. データの準備
# ロングフォーマットへの変換
traits = ['EX', 'OP', 'CO', 'AG']
questions = ['preference', 'believability', 'overall_liking', 'persuasiveness', 'interest', 'click_likelihood']

long_data = []

for trait in traits:
    for q_num, question in enumerate(questions, 1):
        temp_df = data[['ResponseId', f'{trait}_distance_from_max', 'average_distance'] + 
                       [f'Ads_{trait}_{i}' for i in range(1, 7)]].copy()
        temp_df['trait'] = trait
        temp_df['question'] = question
        temp_df['specialized_score'] = temp_df[f'Ads_{trait}_1']
        temp_df['blended_score'] = temp_df[f'Ads_{trait}_{q_num}']
        long_data.append(temp_df)

long_df = pd.concat(long_data, ignore_index=True)

# 相対的選好スコアの計算
long_df['relative_score'] = long_df['blended_score'] - long_df['specialized_score']

# 2. 階層線形モデルの構築
# ヌルモデル
null_model = smf.mixedlm("relative_score ~ 1", data=long_df, groups="ResponseId")
null_fit = null_model.fit()

print("Null Model Results:")
print(null_fit.summary())

# ランダム切片モデル
random_intercept_model = smf.mixedlm("relative_score ~ trait + question", data=long_df, groups="ResponseId")
random_intercept_fit = random_intercept_model.fit()

print("\nRandom Intercept Model Results:")
print(random_intercept_fit.summary())

# ランダム傾きモデル
random_slope_model = smf.mixedlm("relative_score ~ trait + question", 
                                 data=long_df, 
                                 groups="ResponseId", 
                                 re_formula="~trait")
random_slope_fit = random_slope_model.fit()

print("\nRandom Slope Model Results:")
print(random_slope_fit.summary())

# クロスレベル相互作用モデル
# エラーを回避するために、モデル式を修正
interaction_model = smf.mixedlm("relative_score ~ trait + question + trait:average_distance", 
                                data=long_df, 
                                groups="ResponseId", 
                                re_formula="~trait")
interaction_fit = interaction_model.fit()

print("\nInteraction Model Results:")
print(interaction_fit.summary())

# 3. モデル比較
from statsmodels.tools.eval_measures import aic, bic

models = [null_fit, random_intercept_fit, random_slope_fit, interaction_fit]
model_names = ['Null', 'Random Intercept', 'Random Slope', 'Interaction']

comparison_df = pd.DataFrame({
    'Model': model_names,
    'AIC': [model.aic for model in models],
    'BIC': [model.bic for model in models],
    'Log-Likelihood': [model.llf for model in models]
})

print("\nModel Comparison:")
print(comparison_df)

# 4. 結果の可視化
# トレイトごとの相対的選好スコアの分布
plt.figure(figsize=(10, 6))
sns.boxplot(x='trait', y='relative_score', data=long_df)
plt.title('Distribution of Relative Preference Scores by Trait')
plt.show()

# average_distanceと相対的選好スコアの関係
plt.figure(figsize=(10, 6))
sns.scatterplot(x='average_distance', y='relative_score', hue='trait', data=long_df)
plt.title('Relative Preference Score vs Average Distance')
plt.show()

# 5. 予測プロット
# average_distanceの影響を予測
avg_distances = np.linspace(long_df['average_distance'].min(), long_df['average_distance'].max(), 100)
predictions = []

for trait in traits:
    for distance in avg_distances:
        pred_data = pd.DataFrame({
            'trait': [trait],
            'question': ['preference'],
            'average_distance': [distance]
        })
        pred = interaction_fit.predict(pred_data)
        predictions.append({'trait': trait, 'average_distance': distance, 'predicted_score': pred[0]})

pred_df = pd.DataFrame(predictions)

plt.figure(figsize=(10, 6))
for trait in traits:
    trait_data = pred_df[pred_df['trait'] == trait]
    plt.plot(trait_data['average_distance'], trait_data['predicted_score'], label=trait)

plt.xlabel('Average Distance')
plt.ylabel('Predicted Relative Preference Score')
plt.title('Predicted Relative Preference Score vs Average Distance by Trait')
plt.legend()
plt.show()

# 6. 仮説検証
# Blended広告が全体的に好まれているかのt検定
t_stat, p_value = stats.ttest_1samp(long_df['relative_score'], 0)
print("\nOne-sample t-test for overall preference of Blended ads:")
print(f"t-statistic: {t_stat}")
print(f"p-value: {p_value}")

# トレイトごとのBlended広告の選好
for trait in traits:
    trait_scores = long_df[long_df['trait'] == trait]['relative_score']
    t_stat, p_value = stats.ttest_1samp(trait_scores, 0)
    print(f"\nOne-sample t-test for preference of Blended ads ({trait}):")
    print(f"t-statistic: {t_stat}")
    print(f"p-value: {p_value}")

Null Model Results:
           Mixed Linear Model Regression Results
============================================================
Model:            MixedLM Dependent Variable: relative_score
No. Observations: 2160    Method:             REML          
No. Groups:       90      Scale:              228.6478      
Min. group size:  24      Log-Likelihood:     -8994.8627    
Max. group size:  24      Converged:          Yes           
Mean group size:  24.0                                      
------------------------------------------------------------
                  Coef.  Std.Err.   z    P>|z| [0.025 0.975]
------------------------------------------------------------
Intercept         -0.635    0.664 -0.956 0.339 -1.936  0.667
ResponseId Var    30.158    0.402                           
============================================================


Random Intercept Model Results:
                 Mixed Linear Model Regression Results
=======================================================================
Model:                 MixedLM    Dependent Variable:    relative_score
No. Observations:      2160       Method:                REML          
No. Groups:            90         Scale:                 228.6233      
Min. group size:       24         Log-Likelihood:        -8984.2356    
Max. group size:       24         Converged:             Yes           
Mean group size:       24.0                                            
-----------------------------------------------------------------------
                             Coef.  Std.Err.   z    P>|z| [0.025 0.975]
-----------------------------------------------------------------------
Intercept                     1.336    1.135  1.177 0.239 -0.888  3.560
trait[T.CO]                  -0.509    0.920 -0.553 0.580 -2.313  1.294
trait[T.EX]                  -0.919    0.920 -0.998 0.318 -2.722  0.885
trait[T.OP]                  -0.881    0.920 -0.958 0.338 -2.685  0.922
question[T.click_likelihood] -1.583    1.127 -1.405 0.160 -3.792  0.626
question[T.interest]         -1.725    1.127 -1.531 0.126 -3.934  0.484
question[T.overall_liking]   -1.522    1.127 -1.351 0.177 -3.731  0.687
question[T.persuasiveness]   -2.769    1.127 -2.457 0.014 -4.978 -0.561
question[T.preference]       -0.758    1.127 -0.673 0.501 -2.967  1.451
ResponseId Var               30.159    0.402                           
=======================================================================


Random Slope Model Results:
                  Mixed Linear Model Regression Results
=========================================================================
Model:                 MixedLM     Dependent Variable:     relative_score
No. Observations:      2160        Method:                 REML          
No. Groups:            90          Scale:                  142.3142      
Min. group size:       24          Log-Likelihood:         -8699.1813    
Max. group size:       24          Converged:              Yes           
Mean group size:       24.0                                              
-------------------------------------------------------------------------
                               Coef.  Std.Err.   z    P>|z| [0.025 0.975]
-------------------------------------------------------------------------
Intercept                       1.336    1.140  1.171 0.241 -0.899  3.571
trait[T.CO]                    -0.509    1.557 -0.327 0.744 -3.561  2.542
trait[T.EX]                    -0.919    1.150 -0.798 0.425 -3.173  1.336
trait[T.OP]                    -0.881    1.924 -0.458 0.647 -4.651  2.889
question[T.click_likelihood]   -1.583    0.889 -1.781 0.075 -3.326  0.159
question[T.interest]           -1.725    0.889 -1.940 0.052 -3.468  0.018
question[T.overall_liking]     -1.522    0.889 -1.712 0.087 -3.265  0.221
question[T.persuasiveness]     -2.769    0.889 -3.115 0.002 -4.512 -1.027
question[T.preference]         -0.758    0.889 -0.853 0.394 -2.501  0.984
ResponseId Var                 63.658    1.125                           
ResponseId x trait[T.CO] Cov  -53.345    1.421                           
trait[T.CO] Var               170.741    2.808                           
ResponseId x trait[T.EX] Cov  -46.281    1.116                           
trait[T.CO] x trait[T.EX] Cov  40.985    1.554                           
trait[T.EX] Var                71.678    1.533                           
ResponseId x trait[T.OP] Cov  -57.587    1.694                           
trait[T.CO] x trait[T.OP] Cov  63.293    2.528                           
trait[T.EX] x trait[T.OP] Cov  21.314    1.818                           
trait[T.OP] Var               285.549    4.286                           
=========================================================================


Interaction Model Results:
                  Mixed Linear Model Regression Results
==========================================================================
Model:                  MixedLM     Dependent Variable:     relative_score
No. Observations:       2160        Method:                 REML          
No. Groups:             90          Scale:                  142.3136      
Min. group size:        24          Log-Likelihood:         -8688.0438    
Max. group size:        24          Converged:              Yes           
Mean group size:        24.0                                              
--------------------------------------------------------------------------
                               Coef.  Std.Err.   z    P>|z|  [0.025 0.975]
--------------------------------------------------------------------------
Intercept                       4.947    2.715  1.822 0.068  -0.375 10.268
trait[T.CO]                    -5.344    4.208 -1.270 0.204 -13.591  2.903
trait[T.EX]                    -4.523    3.109 -1.455 0.146 -10.616  1.570
trait[T.OP]                    -5.056    5.222 -0.968 0.333 -15.291  5.178
question[T.click_likelihood]   -1.583    0.889 -1.781 0.075  -3.326  0.159
question[T.interest]           -1.725    0.889 -1.940 0.052  -3.468  0.018
question[T.overall_liking]     -1.522    0.889 -1.712 0.087  -3.265  0.221
question[T.persuasiveness]     -2.769    0.889 -3.115 0.002  -4.512 -1.027
question[T.preference]         -0.758    0.889 -0.853 0.394  -2.501  0.984
trait[AG]:average_distance     -5.834    3.985 -1.464 0.143 -13.644  1.977
trait[CO]:average_distance      1.977    5.305  0.373 0.709  -8.421 12.376
trait[EX]:average_distance     -0.010    3.518 -0.003 0.998  -6.906  6.885
trait[OP]:average_distance      0.911    6.927  0.132 0.895 -12.665 14.487
ResponseId Var                 62.552    1.116                            
ResponseId x trait[T.CO] Cov  -51.408    1.410                            
trait[T.CO] Var               169.450    2.807                            
ResponseId x trait[T.EX] Cov  -44.984    1.109                            
trait[T.CO] x trait[T.EX] Cov  38.918    1.547                            
trait[T.EX] Var                70.945    1.532                            
ResponseId x trait[T.OP] Cov  -56.084    1.691                            
trait[T.CO] x trait[T.OP] Cov  61.035    2.532                            
trait[T.EX] x trait[T.OP] Cov  19.404    1.822                            
trait[T.OP] Var               286.544    4.322                            
==========================================================================


Model Comparison:
              Model  AIC  BIC  Log-Likelihood
0              Null  NaN  NaN    -8994.862675
1  Random Intercept  NaN  NaN    -8984.235623
2      Random Slope  NaN  NaN    -8699.181282
3       Interaction  NaN  NaN    -8688.043753

One-sample t-test for overall preference of Blended ads:
t-statistic: -1.834808299174086
p-value: 0.06667161416861433

One-sample t-test for preference of Blended ads (EX):
t-statistic: -1.7082776533603583
p-value: 0.08816058370201699

One-sample t-test for preference of Blended ads (OP):
t-statistic: -1.0950299333662243
p-value: 0.2739925906530151

One-sample t-test for preference of Blended ads (CO):
t-statistic: -0.7823966822943135
p-value: 0.43432506476910926

One-sample t-test for preference of Blended ads (AG):
t-statistic: -0.0998911398122177
p-value: 0.920467892283989

sandbox 4

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from scipy import stats
from sklearn.preprocessing import StandardScaler
from statsmodels.stats.anova import anova_lm
from statsmodels.formula.api import ols

# データの読み込み（既存のデータフレームを 'data' とします）
data = merged_df

# ステップ1: 相対的選好スコアの作成
traits = ['EX', 'OP', 'CO', 'AG']
questions = ['preference', 'believability', 'overall_liking', 'persuasiveness', 'interest', 'click_likelihood']

for trait in traits:
    for i, question in enumerate(questions, 1):
        data[f'{trait}_{question}_relative_score'] = data[f'Ads_{trait}_{i}'] - 50

# ステップ2: 全体的な選好分析
"""
各特性（EX, OP, CO, AG）について、相対的選好スコアの平均値を計算する。
一標本のt検定を使用して、各特性の平均スコアが0と有意に異なるかを検証する。
コメント: この分析により、全体としてBlended広告が好まれているか、
あるいは特定の特性でBlended広告が特に効果的かを把握できます。
"""
def overall_preference_analysis(trait):
    for question in questions:
        score = data[f'{trait}_{question}_relative_score']
        t_stat, p_value = stats.ttest_1samp(score, 0)
        print(f"{trait} - {question}:")
        print(f"Mean score: {score.mean():.2f}")
        print(f"T-statistic: {t_stat:.2f}, p-value: {p_value:.4f}\n")

for trait in traits:
    overall_preference_analysis(trait)

EX - preference:
Mean score: 3.34
T-statistic: 0.95, p-value: 0.3447

EX - believability:
Mean score: 1.11
T-statistic: 0.35, p-value: 0.7277

EX - overall_liking:
Mean score: 2.78
T-statistic: 0.77, p-value: 0.4438

EX - persuasiveness:
Mean score: 0.62
T-statistic: 0.18, p-value: 0.8550

EX - interest:
Mean score: 3.32
T-statistic: 0.92, p-value: 0.3588

EX - click_likelihood:
Mean score: 3.03
T-statistic: 0.82, p-value: 0.4156

OP - preference:
Mean score: -6.06
T-statistic: -1.80, p-value: 0.0755

OP - believability:
Mean score: -4.21
T-statistic: -1.51, p-value: 0.1351

OP - overall_liking:
Mean score: -6.90
T-statistic: -2.01, p-value: 0.0474

OP - persuasiveness:
Mean score: -8.52
T-statistic: -2.70, p-value: 0.0084

OP - interest:
Mean score: -8.23
T-statistic: -2.53, p-value: 0.0131

OP - click_likelihood:
Mean score: -8.04
T-statistic: -2.36, p-value: 0.0206

CO - preference:
Mean score: -1.84
T-statistic: -0.56, p-value: 0.5746

CO - believability:
Mean score: -0.87
T-statistic: -0.29, p-value: 0.7729

CO - overall_liking:
Mean score: -3.64
T-statistic: -1.07, p-value: 0.2865

CO - persuasiveness:
Mean score: -2.84
T-statistic: -0.89, p-value: 0.3756

CO - interest:
Mean score: -2.16
T-statistic: -0.65, p-value: 0.5198

CO - click_likelihood:
Mean score: -3.11
T-statistic: -0.86, p-value: 0.3921

AG - preference:
Mean score: 9.27
T-statistic: 3.34, p-value: 0.0012

AG - believability:
Mean score: 11.71
T-statistic: 4.62, p-value: 0.0000

AG - overall_liking:
Mean score: 9.42
T-statistic: 3.20, p-value: 0.0019

AG - persuasiveness:
Mean score: 7.41
T-statistic: 2.71, p-value: 0.0080

AG - interest:
Mean score: 7.91
T-statistic: 2.79, p-value: 0.0064

AG - click_likelihood:
Mean score: 9.53
T-statistic: 3.13, p-value: 0.0024

# ステップ3: 距離の影響分析
"""
各参加者について、以下の2つの距離を計算する：
a) Blended personaからの距離（既に計算済みのaverage_distance）
b) Single Trait最大スコア（5）からの距離（既に計算済みの{trait}_distance_from_max）
これらの距離と相対的選好スコアの関係を散布図で可視化する。
距離と相対的選好スコアの相関分析を行う。
コメント: この分析により、参加者の特性プロファイルと広告の選好との関係を理解できます。
また、どちらの距離がより強く選好に影響しているかを把握できます。
"""
def distance_analysis(trait):
    for question in questions:
        # Blended personaからの距離との相関
        corr_blended = stats.pearsonr(data['average_distance'], data[f'{trait}_{question}_relative_score'])
        
        # Single Trait最大スコアからの距離との相関
        corr_single = stats.pearsonr(data[f'{trait}_distance_from_max'], data[f'{trait}_{question}_relative_score'])
        
        print(f"{trait} - {question}:")
        print(f"Correlation with Blended distance: r={corr_blended[0]:.2f}, p={corr_blended[1]:.4f}")
        print(f"Correlation with Single Trait distance: r={corr_single[0]:.2f}, p={corr_single[1]:.4f}\n")
        
        # 散布図
        plt.figure(figsize=(12, 5))
        plt.subplot(1, 2, 1)
        sns.scatterplot(x='average_distance', y=f'{trait}_{question}_relative_score', data=data)
        plt.title(f"{trait} - {question} vs Blended Distance")
        plt.subplot(1, 2, 2)
        sns.scatterplot(x=f'{trait}_distance_from_max', y=f'{trait}_{question}_relative_score', data=data)
        plt.title(f"{trait} - {question} vs Single Trait Distance")
        plt.tight_layout()
        plt.show()

for trait in traits:
    distance_analysis(trait)

EX - preference:
Correlation with Blended distance: r=0.23, p=0.0303
Correlation with Single Trait distance: r=0.26, p=0.0151

EX - believability:
Correlation with Blended distance: r=0.21, p=0.0493
Correlation with Single Trait distance: r=0.29, p=0.0063

EX - overall_liking:
Correlation with Blended distance: r=0.24, p=0.0249
Correlation with Single Trait distance: r=0.15, p=0.1455

EX - persuasiveness:
Correlation with Blended distance: r=0.22, p=0.0355
Correlation with Single Trait distance: r=0.25, p=0.0193

EX - interest:
Correlation with Blended distance: r=0.27, p=0.0100
Correlation with Single Trait distance: r=0.24, p=0.0233

EX - click_likelihood:
Correlation with Blended distance: r=0.21, p=0.0466
Correlation with Single Trait distance: r=0.25, p=0.0153

OP - preference:
Correlation with Blended distance: r=0.03, p=0.8057
Correlation with Single Trait distance: r=0.15, p=0.1675

OP - believability:
Correlation with Blended distance: r=0.02, p=0.8770
Correlation with Single Trait distance: r=0.08, p=0.4466

OP - overall_liking:
Correlation with Blended distance: r=-0.01, p=0.9476
Correlation with Single Trait distance: r=0.08, p=0.4338

OP - persuasiveness:
Correlation with Blended distance: r=0.08, p=0.4485
Correlation with Single Trait distance: r=0.14, p=0.1893

OP - interest:
Correlation with Blended distance: r=0.04, p=0.7074
Correlation with Single Trait distance: r=0.10, p=0.3571

OP - click_likelihood:
Correlation with Blended distance: r=0.05, p=0.6276
Correlation with Single Trait distance: r=0.11, p=0.3235

CO - preference:
Correlation with Blended distance: r=-0.00, p=0.9955
Correlation with Single Trait distance: r=0.05, p=0.6300

CO - believability:
Correlation with Blended distance: r=-0.06, p=0.5701
Correlation with Single Trait distance: r=-0.04, p=0.7128

CO - overall_liking:
Correlation with Blended distance: r=0.04, p=0.6959
Correlation with Single Trait distance: r=0.02, p=0.8766

CO - persuasiveness:
Correlation with Blended distance: r=0.05, p=0.6333
Correlation with Single Trait distance: r=0.02, p=0.8212

CO - interest:
Correlation with Blended distance: r=0.04, p=0.6887
Correlation with Single Trait distance: r=0.03, p=0.7436

CO - click_likelihood:
Correlation with Blended distance: r=0.01, p=0.9319
Correlation with Single Trait distance: r=0.04, p=0.7276

AG - preference:
Correlation with Blended distance: r=0.04, p=0.7425
Correlation with Single Trait distance: r=-0.01, p=0.9587

AG - believability:
Correlation with Blended distance: r=-0.05, p=0.6537
Correlation with Single Trait distance: r=-0.10, p=0.3446

AG - overall_liking:
Correlation with Blended distance: r=-0.06, p=0.6058
Correlation with Single Trait distance: r=-0.01, p=0.9320

AG - persuasiveness:
Correlation with Blended distance: r=0.01, p=0.9442
Correlation with Single Trait distance: r=0.03, p=0.8143

AG - interest:
Correlation with Blended distance: r=-0.03, p=0.7669
Correlation with Single Trait distance: r=0.03, p=0.7835

AG - click_likelihood:
Correlation with Blended distance: r=-0.02, p=0.8280
Correlation with Single Trait distance: r=0.03, p=0.7614

# ステップ4: 広告効果の多面的分析
# ステップ2と3の分析を各質問について繰り返す（上記のコードで既に実装済み）
"""
6つの質問（好ましさ、信頼性、全体的な好み、説得力、興味、クリック可能性）それぞれについて、
ステップ2と3の分析を繰り返す。
コメント: この分析により、Blended広告が特に効果的な側面
（例：信頼性は高いが、クリック可能性は低いなど）を特定できます。
"""

def multifaceted_ad_effect_analysis(trait):
    for question in questions:
        print(f"\n{trait} - {question} 分析結果:")
        
        # 1. 全体的な選好分析（ステップ2の一部）
        score = data[f'{trait}_{question}_relative_score']
        t_stat, p_value = stats.ttest_1samp(score, 0)
        print(f"全体的な選好:")
        print(f"平均スコア: {score.mean():.2f}")
        print(f"T統計量: {t_stat:.2f}, p値: {p_value:.4f}")
        
        # 2. 距離の影響分析（ステップ3の一部）
        corr_blended = stats.pearsonr(data['average_distance'], score)
        corr_single = stats.pearsonr(data[f'{trait}_distance_from_max'], score)
        
        print("\n距離との相関:")
        print(f"Blended距離との相関: r={corr_blended[0]:.2f}, p={corr_blended[1]:.4f}")
        print(f"Single Trait距離との相関: r={corr_single[0]:.2f}, p={corr_single[1]:.4f}")
        
        # 3. 散布図の作成
        plt.figure(figsize=(12, 5))
        plt.subplot(1, 2, 1)
        sns.scatterplot(x='average_distance', y=f'{trait}_{question}_relative_score', data=data)
        plt.title(f"{trait} - {question} vs Blended Distance")
        plt.subplot(1, 2, 2)
        sns.scatterplot(x=f'{trait}_distance_from_max', y=f'{trait}_{question}_relative_score', data=data)
        plt.title(f"{trait} - {question} vs Single Trait Distance")
        plt.tight_layout()
        plt.show()
        
        # 4. 回帰分析
        model = smf.ols(f'{trait}_{question}_relative_score ~ average_distance + {trait}_distance_from_max', data=data)
        results = model.fit()
        print("\n回帰分析結果:")
        print(results.summary())
        
        print("\n" + "="*50)

# 各特性について多面的分析を実行
for trait in traits:
    multifaceted_ad_effect_analysis(trait)

# 質問間の相関分析
def question_correlation_analysis(trait):
    correlation_matrix = data[[f'{trait}_{q}_relative_score' for q in questions]].corr()
    plt.figure(figsize=(10, 8))
    sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm', vmin=-1, vmax=1)
    plt.title(f'{trait} - 質問間の相関')
    plt.tight_layout()
    plt.show()

for trait in traits:
    question_correlation_analysis(trait)

# 質問間の比較
def question_comparison(trait):
    scores = [data[f'{trait}_{q}_relative_score'] for q in questions]
    f_statistic, p_value = stats.f_oneway(*scores)
    print(f"\n{trait} - 質問間の比較 (ANOVA):")
    print(f"F統計量: {f_statistic:.2f}, p値: {p_value:.4f}")
    
    plt.figure(figsize=(10, 6))
    sns.boxplot(data=pd.melt(data[[f'{trait}_{q}_relative_score' for q in questions]], 
                             var_name='Question', value_name='Score'))
    plt.title(f'{trait} - 質問ごとの相対的選好スコア分布')
    plt.xticks(rotation=45)
    plt.tight_layout()
    plt.show()

for trait in traits:
    question_comparison(trait)

EX - preference 分析結果:
全体的な選好:
平均スコア: 3.34
T統計量: 0.95, p値: 0.3447

距離との相関:
Blended距離との相関: r=0.23, p=0.0303
Single Trait距離との相関: r=0.26, p=0.0151

回帰分析結果:
                                 OLS Regression Results                                 
========================================================================================
Dep. Variable:     EX_preference_relative_score   R-squared:                       0.103
Model:                                      OLS   Adj. R-squared:                  0.082
Method:                           Least Squares   F-statistic:                     4.986
Date:                          Thu, 25 Jul 2024   Prob (F-statistic):            0.00891
Time:                                  08:00:17   Log-Likelihood:                -438.09
No. Observations:                            90   AIC:                             882.2
Df Residuals:                                87   BIC:                             889.7
Df Model:                                     2                                         
Covariance Type:                      nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept              -30.0137     11.175     -2.686      0.009     -52.226      -7.802
average_distance        26.4893     13.872      1.910      0.059      -1.083      54.062
EX_distance_from_max     8.8326      3.986      2.216      0.029       0.909      16.756
==============================================================================
Omnibus:                       21.498   Durbin-Watson:                   1.739
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                4.968
Skew:                          -0.102   Prob(JB):                       0.0834
Kurtosis:                       1.867   Cond. No.                         11.5
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

EX - believability 分析結果:
全体的な選好:
平均スコア: 1.11
T統計量: 0.35, p値: 0.7277

距離との相関:
Blended距離との相関: r=0.21, p=0.0493
Single Trait距離との相関: r=0.29, p=0.0063

回帰分析結果:
                                   OLS Regression Results                                  
===========================================================================================
Dep. Variable:     EX_believability_relative_score   R-squared:                       0.110
Model:                                         OLS   Adj. R-squared:                  0.090
Method:                              Least Squares   F-statistic:                     5.384
Date:                             Thu, 25 Jul 2024   Prob (F-statistic):            0.00624
Time:                                     08:00:18   Log-Likelihood:                -428.58
No. Observations:                               90   AIC:                             863.2
Df Residuals:                                   87   BIC:                             870.7
Df Model:                                        2                                         
Covariance Type:                         nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept              -29.3902     10.054     -2.923      0.004     -49.374      -9.407
average_distance        20.8114     12.480      1.668      0.099      -3.995      45.618
EX_distance_from_max     9.1750      3.586      2.558      0.012       2.047      16.303
==============================================================================
Omnibus:                        9.881   Durbin-Watson:                   1.766
Prob(Omnibus):                  0.007   Jarque-Bera (JB):                3.406
Skew:                           0.022   Prob(JB):                        0.182
Kurtosis:                       2.048   Cond. No.                         11.5
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

EX - overall_liking 分析結果:
全体的な選好:
平均スコア: 2.78
T統計量: 0.77, p値: 0.4438

距離との相関:
Blended距離との相関: r=0.24, p=0.0249
Single Trait距離との相関: r=0.15, p=0.1455

回帰分析結果:
                                   OLS Regression Results                                   
============================================================================================
Dep. Variable:     EX_overall_liking_relative_score   R-squared:                       0.071
Model:                                          OLS   Adj. R-squared:                  0.049
Method:                               Least Squares   F-statistic:                     3.312
Date:                              Thu, 25 Jul 2024   Prob (F-statistic):             0.0411
Time:                                      08:00:18   Log-Likelihood:                -441.95
No. Observations:                                90   AIC:                             889.9
Df Residuals:                                    87   BIC:                             897.4
Df Model:                                         2                                         
Covariance Type:                          nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept              -25.4264     11.664     -2.180      0.032     -48.611      -2.242
average_distance        30.3177     14.480      2.094      0.039       1.538      59.097
EX_distance_from_max     4.9147      4.161      1.181      0.241      -3.355      13.185
==============================================================================
Omnibus:                       66.369   Durbin-Watson:                   1.782
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.181
Skew:                          -0.103   Prob(JB):                       0.0276
Kurtosis:                       1.632   Cond. No.                         11.5
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

EX - persuasiveness 分析結果:
全体的な選好:
平均スコア: 0.62
T統計量: 0.18, p値: 0.8550

距離との相関:
Blended距離との相関: r=0.22, p=0.0355
Single Trait距離との相関: r=0.25, p=0.0193

回帰分析結果:
                                   OLS Regression Results                                   
============================================================================================
Dep. Variable:     EX_persuasiveness_relative_score   R-squared:                       0.096
Model:                                          OLS   Adj. R-squared:                  0.075
Method:                               Least Squares   F-statistic:                     4.631
Date:                              Thu, 25 Jul 2024   Prob (F-statistic):             0.0123
Time:                                      08:00:19   Log-Likelihood:                -435.15
No. Observations:                                90   AIC:                             876.3
Df Residuals:                                    87   BIC:                             883.8
Df Model:                                         2                                         
Covariance Type:                          nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept              -30.5094     10.815     -2.821      0.006     -52.006      -9.013
average_distance        24.8439     13.426      1.850      0.068      -1.841      51.529
EX_distance_from_max     8.2035      3.858      2.126      0.036       0.535      15.871
==============================================================================
Omnibus:                       25.099   Durbin-Watson:                   1.671
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.164
Skew:                           0.047   Prob(JB):                       0.0756
Kurtosis:                       1.830   Cond. No.                         11.5
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

EX - interest 分析結果:
全体的な選好:
平均スコア: 3.32
T統計量: 0.92, p値: 0.3588

距離との相関:
Blended距離との相関: r=0.27, p=0.0100
Single Trait距離との相関: r=0.24, p=0.0233

回帰分析結果:
                                OLS Regression Results                                
======================================================================================
Dep. Variable:     EX_interest_relative_score   R-squared:                       0.114
Model:                                    OLS   Adj. R-squared:                  0.094
Method:                         Least Squares   F-statistic:                     5.592
Date:                        Thu, 25 Jul 2024   Prob (F-statistic):            0.00519
Time:                                08:00:19   Log-Likelihood:                -439.57
No. Observations:                          90   AIC:                             885.1
Df Residuals:                              87   BIC:                             892.6
Df Model:                                   2                                         
Covariance Type:                    nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept              -32.8947     11.360     -2.896      0.005     -55.473     -10.316
average_distance        33.3058     14.101      2.362      0.020       5.278      61.333
EX_distance_from_max     8.1241      4.052      2.005      0.048       0.070      16.178
==============================================================================
Omnibus:                       24.032   Durbin-Watson:                   1.687
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.290
Skew:                          -0.133   Prob(JB):                       0.0710
Kurtosis:                       1.843   Cond. No.                         11.5
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

EX - click_likelihood 分析結果:
全体的な選好:
平均スコア: 3.03
T統計量: 0.82, p値: 0.4156

距離との相関:
Blended距離との相関: r=0.21, p=0.0466
Single Trait距離との相関: r=0.25, p=0.0153

回帰分析結果:
                                    OLS Regression Results                                    
==============================================================================================
Dep. Variable:     EX_click_likelihood_relative_score   R-squared:                       0.096
Model:                                            OLS   Adj. R-squared:                  0.075
Method:                                 Least Squares   F-statistic:                     4.612
Date:                                Thu, 25 Jul 2024   Prob (F-statistic):             0.0125
Time:                                        08:00:20   Log-Likelihood:                -443.11
No. Observations:                                  90   AIC:                             892.2
Df Residuals:                                      87   BIC:                             899.7
Df Model:                                           2                                         
Covariance Type:                            nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept              -30.6494     11.816     -2.594      0.011     -54.135      -7.163
average_distance        25.2738     14.668      1.723      0.088      -3.880      54.428
EX_distance_from_max     9.3934      4.215      2.229      0.028       1.016      17.771
==============================================================================
Omnibus:                       27.502   Durbin-Watson:                   1.736
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.457
Skew:                          -0.101   Prob(JB):                       0.0653
Kurtosis:                       1.811   Cond. No.                         11.5
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

OP - preference 分析結果:
全体的な選好:
平均スコア: -6.06
T統計量: -1.80, p値: 0.0755

距離との相関:
Blended距離との相関: r=0.03, p=0.8057
Single Trait距離との相関: r=0.15, p=0.1675

回帰分析結果:
                                 OLS Regression Results                                 
========================================================================================
Dep. Variable:     OP_preference_relative_score   R-squared:                       0.022
Model:                                      OLS   Adj. R-squared:                 -0.001
Method:                           Least Squares   F-statistic:                    0.9581
Date:                          Thu, 25 Jul 2024   Prob (F-statistic):              0.388
Time:                                  08:00:20   Log-Likelihood:                -437.98
No. Observations:                            90   AIC:                             882.0
Df Residuals:                                87   BIC:                             889.5
Df Model:                                     2                                         
Covariance Type:                      nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept              -13.6485      9.981     -1.367      0.175     -33.487       6.191
average_distance        -0.5649     14.018     -0.040      0.968     -28.427      27.297
OP_distance_from_max     6.4303      4.722      1.362      0.177      -2.954      15.815
==============================================================================
Omnibus:                       38.063   Durbin-Watson:                   1.888
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                6.967
Skew:                           0.267   Prob(JB):                       0.0307
Kurtosis:                       1.746   Cond. No.                         8.67
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

OP - believability 分析結果:
全体的な選好:
平均スコア: -4.21
T統計量: -1.51, p値: 0.1351

距離との相関:
Blended距離との相関: r=0.02, p=0.8770
Single Trait距離との相関: r=0.08, p=0.4466

回帰分析結果:
                                   OLS Regression Results                                  
===========================================================================================
Dep. Variable:     OP_believability_relative_score   R-squared:                       0.007
Model:                                         OLS   Adj. R-squared:                 -0.016
Method:                              Least Squares   F-statistic:                    0.2889
Date:                             Thu, 25 Jul 2024   Prob (F-statistic):              0.750
Time:                                     08:00:21   Log-Likelihood:                -421.82
No. Observations:                               90   AIC:                             849.6
Df Residuals:                                   87   BIC:                             857.1
Df Model:                                        2                                         
Covariance Type:                         nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept               -7.8151      8.341     -0.937      0.351     -24.393       8.763
average_distance        -0.0367     11.713     -0.003      0.998     -23.318      23.245
OP_distance_from_max     2.9361      3.945      0.744      0.459      -4.906      10.778
==============================================================================
Omnibus:                        3.824   Durbin-Watson:                   2.223
Prob(Omnibus):                  0.148   Jarque-Bera (JB):                2.089
Skew:                           0.064   Prob(JB):                        0.352
Kurtosis:                       2.265   Cond. No.                         8.67
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

OP - overall_liking 分析結果:
全体的な選好:
平均スコア: -6.90
T統計量: -2.01, p値: 0.0474

距離との相関:
Blended距離との相関: r=-0.01, p=0.9476
Single Trait距離との相関: r=0.08, p=0.4338

回帰分析結果:
                                   OLS Regression Results                                   
============================================================================================
Dep. Variable:     OP_overall_liking_relative_score   R-squared:                       0.008
Model:                                          OLS   Adj. R-squared:                 -0.015
Method:                               Least Squares   F-statistic:                    0.3331
Date:                              Thu, 25 Jul 2024   Prob (F-statistic):              0.718
Time:                                      08:00:21   Log-Likelihood:                -440.33
No. Observations:                                90   AIC:                             886.7
Df Residuals:                                    87   BIC:                             894.2
Df Model:                                         2                                         
Covariance Type:                          nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept               -9.6923     10.246     -0.946      0.347     -30.057      10.672
average_distance        -3.3568     14.389     -0.233      0.816     -31.956      25.243
OP_distance_from_max     3.9428      4.847      0.814      0.418      -5.690      13.576
==============================================================================
Omnibus:                       46.816   Durbin-Watson:                   1.936
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.457
Skew:                           0.276   Prob(JB):                       0.0240
Kurtosis:                       1.702   Cond. No.                         8.67
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

OP - persuasiveness 分析結果:
全体的な選好:
平均スコア: -8.52
T統計量: -2.70, p値: 0.0084

距離との相関:
Blended距離との相関: r=0.08, p=0.4485
Single Trait距離との相関: r=0.14, p=0.1893

回帰分析結果:
                                   OLS Regression Results                                   
============================================================================================
Dep. Variable:     OP_persuasiveness_relative_score   R-squared:                       0.022
Model:                                          OLS   Adj. R-squared:                 -0.000
Method:                               Least Squares   F-statistic:                    0.9930
Date:                              Thu, 25 Jul 2024   Prob (F-statistic):              0.375
Time:                                      08:00:22   Log-Likelihood:                -432.26
No. Observations:                                90   AIC:                             870.5
Df Residuals:                                    87   BIC:                             878.0
Df Model:                                         2                                         
Covariance Type:                          nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept              -19.0802      9.366     -2.037      0.045     -37.697      -0.464
average_distance         6.5826     13.154      0.500      0.618     -19.562      32.727
OP_distance_from_max     5.2490      4.431      1.185      0.239      -3.557      14.055
==============================================================================
Omnibus:                       13.810   Durbin-Watson:                   1.966
Prob(Omnibus):                  0.001   Jarque-Bera (JB):                5.697
Skew:                           0.367   Prob(JB):                       0.0579
Kurtosis:                       2.010   Cond. No.                         8.67
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

OP - interest 分析結果:
全体的な選好:
平均スコア: -8.23
T統計量: -2.53, p値: 0.0131

距離との相関:
Blended距離との相関: r=0.04, p=0.7074
Single Trait距離との相関: r=0.10, p=0.3571

回帰分析結果:
                                OLS Regression Results                                
======================================================================================
Dep. Variable:     OP_interest_relative_score   R-squared:                       0.010
Model:                                    OLS   Adj. R-squared:                 -0.013
Method:                         Least Squares   F-statistic:                    0.4416
Date:                        Thu, 25 Jul 2024   Prob (F-statistic):              0.644
Time:                                08:00:22   Log-Likelihood:                -435.37
No. Observations:                          90   AIC:                             876.7
Df Residuals:                              87   BIC:                             884.2
Df Model:                                   2                                         
Covariance Type:                    nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept              -14.7051      9.696     -1.517      0.133     -33.977       4.567
average_distance         2.5724     13.617      0.189      0.851     -24.493      29.638
OP_distance_from_max     3.9504      4.587      0.861      0.391      -5.166      13.067
==============================================================================
Omnibus:                       17.137   Durbin-Watson:                   2.016
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                6.616
Skew:                           0.413   Prob(JB):                       0.0366
Kurtosis:                       1.959   Cond. No.                         8.67
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

OP - click_likelihood 分析結果:
全体的な選好:
平均スコア: -8.04
T統計量: -2.36, p値: 0.0206

距離との相関:
Blended距離との相関: r=0.05, p=0.6276
Single Trait距離との相関: r=0.11, p=0.3235

回帰分析結果:
                                    OLS Regression Results                                    
==============================================================================================
Dep. Variable:     OP_click_likelihood_relative_score   R-squared:                       0.012
Model:                                            OLS   Adj. R-squared:                 -0.011
Method:                                 Least Squares   F-statistic:                    0.5291
Date:                                Thu, 25 Jul 2024   Prob (F-statistic):              0.591
Time:                                        08:00:23   Log-Likelihood:                -439.62
No. Observations:                                  90   AIC:                             885.2
Df Residuals:                                      87   BIC:                             892.7
Df Model:                                           2                                         
Covariance Type:                            nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept              -15.9680     10.164     -1.571      0.120     -36.171       4.235
average_distance         4.1037     14.275      0.287      0.774     -24.269      32.476
OP_distance_from_max     4.3584      4.808      0.906      0.367      -5.198      13.915
==============================================================================
Omnibus:                       38.012   Durbin-Watson:                   1.871
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.430
Skew:                           0.323   Prob(JB):                       0.0244
Kurtosis:                       1.749   Cond. No.                         8.67
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

CO - preference 分析結果:
全体的な選好:
平均スコア: -1.84
T統計量: -0.56, p値: 0.5746

距離との相関:
Blended距離との相関: r=-0.00, p=0.9955
Single Trait距離との相関: r=0.05, p=0.6300

回帰分析結果:
                                 OLS Regression Results                                 
========================================================================================
Dep. Variable:     CO_preference_relative_score   R-squared:                       0.003
Model:                                      OLS   Adj. R-squared:                 -0.020
Method:                           Least Squares   F-statistic:                    0.1193
Date:                          Thu, 25 Jul 2024   Prob (F-statistic):              0.888
Time:                                  08:00:23   Log-Likelihood:                -436.31
No. Observations:                            90   AIC:                             878.6
Df Residuals:                                87   BIC:                             886.1
Df Model:                                     2                                         
Covariance Type:                      nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept               -3.4064      9.550     -0.357      0.722     -22.388      15.575
average_distance        -1.1954     13.654     -0.088      0.930     -28.333      25.942
CO_distance_from_max     2.2157      4.536      0.488      0.626      -6.800      11.231
==============================================================================
Omnibus:                       40.136   Durbin-Watson:                   2.122
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                6.125
Skew:                           0.064   Prob(JB):                       0.0468
Kurtosis:                       1.728   Cond. No.                         8.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

CO - believability 分析結果:
全体的な選好:
平均スコア: -0.87
T統計量: -0.29, p値: 0.7729

距離との相関:
Blended距離との相関: r=-0.06, p=0.5701
Single Trait距離との相関: r=-0.04, p=0.7128

回帰分析結果:
                                   OLS Regression Results                                  
===========================================================================================
Dep. Variable:     CO_believability_relative_score   R-squared:                       0.005
Model:                                         OLS   Adj. R-squared:                 -0.018
Method:                              Least Squares   F-statistic:                    0.1990
Date:                             Thu, 25 Jul 2024   Prob (F-statistic):              0.820
Time:                                     08:00:24   Log-Likelihood:                -428.20
No. Observations:                               90   AIC:                             862.4
Df Residuals:                                   87   BIC:                             869.9
Df Model:                                        2                                         
Covariance Type:                         nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept                4.2826      8.727      0.491      0.625     -13.063      21.628
average_distance        -6.3952     12.476     -0.513      0.610     -31.193      18.403
CO_distance_from_max    -1.1462      4.145     -0.277      0.783      -9.384       7.092
==============================================================================
Omnibus:                        6.127   Durbin-Watson:                   2.068
Prob(Omnibus):                  0.047   Jarque-Bera (JB):                2.687
Skew:                           0.052   Prob(JB):                        0.261
Kurtosis:                       2.160   Cond. No.                         8.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

CO - overall_liking 分析結果:
全体的な選好:
平均スコア: -3.64
T統計量: -1.07, p値: 0.2865

距離との相関:
Blended距離との相関: r=0.04, p=0.6959
Single Trait距離との相関: r=0.02, p=0.8766

回帰分析結果:
                                   OLS Regression Results                                   
============================================================================================
Dep. Variable:     CO_overall_liking_relative_score   R-squared:                       0.002
Model:                                          OLS   Adj. R-squared:                 -0.021
Method:                               Least Squares   F-statistic:                   0.08013
Date:                              Thu, 25 Jul 2024   Prob (F-statistic):              0.923
Time:                                      08:00:24   Log-Likelihood:                -439.72
No. Observations:                                90   AIC:                             885.4
Df Residuals:                                    87   BIC:                             892.9
Df Model:                                         2                                         
Covariance Type:                          nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept               -7.3289      9.919     -0.739      0.462     -27.044      12.386
average_distance         5.2343     14.181      0.369      0.713     -22.952      33.420
CO_distance_from_max     0.4279      4.711      0.091      0.928      -8.936       9.791
==============================================================================
Omnibus:                       73.154   Durbin-Watson:                   1.993
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                7.397
Skew:                           0.115   Prob(JB):                       0.0248
Kurtosis:                       1.615   Cond. No.                         8.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

CO - persuasiveness 分析結果:
全体的な選好:
平均スコア: -2.84
T統計量: -0.89, p値: 0.3756

距離との相関:
Blended距離との相関: r=0.05, p=0.6333
Single Trait距離との相関: r=0.02, p=0.8212

回帰分析結果:
                                   OLS Regression Results                                   
============================================================================================
Dep. Variable:     CO_persuasiveness_relative_score   R-squared:                       0.003
Model:                                          OLS   Adj. R-squared:                 -0.020
Method:                               Least Squares   F-statistic:                    0.1242
Date:                              Thu, 25 Jul 2024   Prob (F-statistic):              0.883
Time:                                      08:00:25   Log-Likelihood:                -434.09
No. Observations:                                90   AIC:                             874.2
Df Residuals:                                    87   BIC:                             881.7
Df Model:                                         2                                         
Covariance Type:                          nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept               -7.1891      9.317     -0.772      0.442     -25.707      11.329
average_distance         5.9207     13.320      0.444      0.658     -20.555      32.396
CO_distance_from_max     0.6544      4.425      0.148      0.883      -8.141       9.450
==============================================================================
Omnibus:                       25.081   Durbin-Watson:                   2.014
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.246
Skew:                           0.091   Prob(JB):                       0.0726
Kurtosis:                       1.831   Cond. No.                         8.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

CO - interest 分析結果:
全体的な選好:
平均スコア: -2.16
T統計量: -0.65, p値: 0.5198

距離との相関:
Blended距離との相関: r=0.04, p=0.6887
Single Trait距離との相関: r=0.03, p=0.7436

回帰分析結果:
                                OLS Regression Results                                
======================================================================================
Dep. Variable:     CO_interest_relative_score   R-squared:                       0.003
Model:                                    OLS   Adj. R-squared:                 -0.020
Method:                         Least Squares   F-statistic:                    0.1145
Date:                        Thu, 25 Jul 2024   Prob (F-statistic):              0.892
Time:                                08:00:26   Log-Likelihood:                -438.00
No. Observations:                          90   AIC:                             882.0
Df Residuals:                              87   BIC:                             889.5
Df Model:                                   2                                         
Covariance Type:                    nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept               -6.4330      9.731     -0.661      0.510     -25.775      12.908
average_distance         4.8698     13.912      0.350      0.727     -22.782      32.522
CO_distance_from_max     1.2159      4.622      0.263      0.793      -7.970      10.402
==============================================================================
Omnibus:                       32.478   Durbin-Watson:                   2.024
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                5.788
Skew:                           0.100   Prob(JB):                       0.0554
Kurtosis:                       1.774   Cond. No.                         8.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

CO - click_likelihood 分析結果:
全体的な選好:
平均スコア: -3.11
T統計量: -0.86, p値: 0.3921

距離との相関:
Blended距離との相関: r=0.01, p=0.9319
Single Trait距離との相関: r=0.04, p=0.7276

回帰分析結果:
                                    OLS Regression Results                                    
==============================================================================================
Dep. Variable:     CO_click_likelihood_relative_score   R-squared:                       0.001
Model:                                            OLS   Adj. R-squared:                 -0.022
Method:                                 Least Squares   F-statistic:                   0.06073
Date:                                Thu, 25 Jul 2024   Prob (F-statistic):              0.941
Time:                                        08:00:27   Log-Likelihood:                -445.36
No. Observations:                                  90   AIC:                             896.7
Df Residuals:                                      87   BIC:                             904.2
Df Model:                                           2                                         
Covariance Type:                            nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept               -5.1280     10.560     -0.486      0.628     -26.117      15.861
average_distance         0.4140     15.097      0.027      0.978     -29.593      30.421
CO_distance_from_max     1.6947      5.015      0.338      0.736      -8.274      11.663
==============================================================================
Omnibus:                      140.150   Durbin-Watson:                   2.006
Prob(Omnibus):                  0.000   Jarque-Bera (JB):                8.361
Skew:                           0.089   Prob(JB):                       0.0153
Kurtosis:                       1.517   Cond. No.                         8.00
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

AG - preference 分析結果:
全体的な選好:
平均スコア: 9.27
T統計量: 3.34, p値: 0.0012

距離との相関:
Blended距離との相関: r=0.04, p=0.7425
Single Trait距離との相関: r=-0.01, p=0.9587

回帰分析結果:
                                 OLS Regression Results                                 
========================================================================================
Dep. Variable:     AG_preference_relative_score   R-squared:                       0.001
Model:                                      OLS   Adj. R-squared:                 -0.022
Method:                           Least Squares   F-statistic:                   0.06090
Date:                          Thu, 25 Jul 2024   Prob (F-statistic):              0.941
Time:                                  08:00:27   Log-Likelihood:                -421.60
No. Observations:                            90   AIC:                             849.2
Df Residuals:                                87   BIC:                             856.7
Df Model:                                     2                                         
Covariance Type:                      nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept                7.3061      8.178      0.893      0.374      -8.948      23.561
average_distance         4.0273     11.668      0.345      0.731     -19.164      27.218
AG_distance_from_max    -0.4921      4.104     -0.120      0.905      -8.650       7.666
==============================================================================
Omnibus:                        5.821   Durbin-Watson:                   1.892
Prob(Omnibus):                  0.054   Jarque-Bera (JB):                5.091
Skew:                          -0.497   Prob(JB):                       0.0784
Kurtosis:                       2.394   Cond. No.                         8.10
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

AG - believability 分析結果:
全体的な選好:
平均スコア: 11.71
T統計量: 4.62, p値: 0.0000

距離との相関:
Blended距離との相関: r=-0.05, p=0.6537
Single Trait距離との相関: r=-0.10, p=0.3446

回帰分析結果:
                                   OLS Regression Results                                  
===========================================================================================
Dep. Variable:     AG_believability_relative_score   R-squared:                       0.011
Model:                                         OLS   Adj. R-squared:                 -0.012
Method:                              Least Squares   F-statistic:                    0.4819
Date:                             Thu, 25 Jul 2024   Prob (F-statistic):              0.619
Time:                                     08:00:27   Log-Likelihood:                -412.84
No. Observations:                               90   AIC:                             831.7
Df Residuals:                                   87   BIC:                             839.2
Df Model:                                        2                                         
Covariance Type:                         nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept               16.9634      7.419      2.286      0.025       2.217      31.710
average_distance        -2.8071     10.585     -0.265      0.791     -23.846      18.232
AG_distance_from_max    -3.2499      3.724     -0.873      0.385     -10.651       4.151
==============================================================================
Omnibus:                        3.126   Durbin-Watson:                   1.855
Prob(Omnibus):                  0.209   Jarque-Bera (JB):                3.118
Skew:                          -0.426   Prob(JB):                        0.210
Kurtosis:                       2.675   Cond. No.                         8.10
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

AG - overall_liking 分析結果:
全体的な選好:
平均スコア: 9.42
T統計量: 3.20, p値: 0.0019

距離との相関:
Blended距離との相関: r=-0.06, p=0.6058
Single Trait距離との相関: r=-0.01, p=0.9320

回帰分析結果:
                                   OLS Regression Results                                   
============================================================================================
Dep. Variable:     AG_overall_liking_relative_score   R-squared:                       0.003
Model:                                          OLS   Adj. R-squared:                 -0.020
Method:                               Least Squares   F-statistic:                    0.1328
Date:                              Thu, 25 Jul 2024   Prob (F-statistic):              0.876
Time:                                      08:00:28   Log-Likelihood:                -426.72
No. Observations:                                90   AIC:                             859.4
Df Residuals:                                    87   BIC:                             866.9
Df Model:                                         2                                         
Covariance Type:                          nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept               13.2211      8.657      1.527      0.130      -3.986      30.428
average_distance        -6.2777     12.351     -0.508      0.613     -30.827      18.272
AG_distance_from_max     0.0804      4.345      0.019      0.985      -8.555       8.716
==============================================================================
Omnibus:                        6.489   Durbin-Watson:                   1.944
Prob(Omnibus):                  0.039   Jarque-Bera (JB):                4.136
Skew:                          -0.358   Prob(JB):                        0.126
Kurtosis:                       2.231   Cond. No.                         8.10
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

AG - persuasiveness 分析結果:
全体的な選好:
平均スコア: 7.41
T統計量: 2.71, p値: 0.0080

距離との相関:
Blended距離との相関: r=0.01, p=0.9442
Single Trait距離との相関: r=0.03, p=0.8143

回帰分析結果:
                                   OLS Regression Results                                   
============================================================================================
Dep. Variable:     AG_persuasiveness_relative_score   R-squared:                       0.001
Model:                                          OLS   Adj. R-squared:                 -0.022
Method:                               Least Squares   F-statistic:                   0.02770
Date:                              Thu, 25 Jul 2024   Prob (F-statistic):              0.973
Time:                                      08:00:29   Log-Likelihood:                -420.08
No. Observations:                                90   AIC:                             846.2
Df Residuals:                                    87   BIC:                             853.7
Df Model:                                         2                                         
Covariance Type:                          nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept                6.2646      8.041      0.779      0.438      -9.718      22.247
average_distance         0.2674     11.472      0.023      0.981     -22.535      23.070
AG_distance_from_max     0.9071      4.036      0.225      0.823      -7.114       8.928
==============================================================================
Omnibus:                        3.105   Durbin-Watson:                   1.897
Prob(Omnibus):                  0.212   Jarque-Bera (JB):                2.529
Skew:                          -0.287   Prob(JB):                        0.282
Kurtosis:                       2.412   Cond. No.                         8.10
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

AG - interest 分析結果:
全体的な選好:
平均スコア: 7.91
T統計量: 2.79, p値: 0.0064

距離との相関:
Blended距離との相関: r=-0.03, p=0.7669
Single Trait距離との相関: r=0.03, p=0.7835

回帰分析結果:
                                OLS Regression Results                                
======================================================================================
Dep. Variable:     AG_interest_relative_score   R-squared:                       0.002
Model:                                    OLS   Adj. R-squared:                 -0.021
Method:                         Least Squares   F-statistic:                    0.1017
Date:                        Thu, 25 Jul 2024   Prob (F-statistic):              0.903
Time:                                08:00:29   Log-Likelihood:                -423.28
No. Observations:                          90   AIC:                             852.6
Df Residuals:                              87   BIC:                             860.1
Df Model:                                   2                                         
Covariance Type:                    nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept                9.0064      8.332      1.081      0.283      -7.554      25.566
average_distance        -4.2575     11.887     -0.358      0.721     -27.884      19.369
AG_distance_from_max     1.4240      4.181      0.341      0.734      -6.887       9.735
==============================================================================
Omnibus:                        7.231   Durbin-Watson:                   1.966
Prob(Omnibus):                  0.027   Jarque-Bera (JB):                3.538
Skew:                          -0.237   Prob(JB):                        0.171
Kurtosis:                       2.152   Cond. No.                         8.10
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

AG - click_likelihood 分析結果:
全体的な選好:
平均スコア: 9.53
T統計量: 3.13, p値: 0.0024

距離との相関:
Blended距離との相関: r=-0.02, p=0.8280
Single Trait距離との相関: r=0.03, p=0.7614

回帰分析結果:
                                    OLS Regression Results                                    
==============================================================================================
Dep. Variable:     AG_click_likelihood_relative_score   R-squared:                       0.002
Model:                                            OLS   Adj. R-squared:                 -0.021
Method:                                 Least Squares   F-statistic:                   0.08606
Date:                                Thu, 25 Jul 2024   Prob (F-statistic):              0.918
Time:                                        08:00:30   Log-Likelihood:                -429.82
No. Observations:                                  90   AIC:                             865.6
Df Residuals:                                      87   BIC:                             873.1
Df Model:                                           2                                         
Covariance Type:                            nonrobust                                         
========================================================================================
                           coef    std err          t      P>|t|      [0.025      0.975]
----------------------------------------------------------------------------------------
Intercept               10.0560      8.959      1.122      0.265      -7.752      27.864
average_distance        -3.6231     12.783     -0.283      0.778     -29.030      21.784
AG_distance_from_max     1.5903      4.497      0.354      0.724      -7.347      10.528
==============================================================================
Omnibus:                        9.406   Durbin-Watson:                   1.786
Prob(Omnibus):                  0.009   Jarque-Bera (JB):                4.671
Skew:                          -0.337   Prob(JB):                       0.0968
Kurtosis:                       2.111   Cond. No.                         8.10
==============================================================================

Notes:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

==================================================

EX - 質問間の比較 (ANOVA):
F統計量: 0.12, p値: 0.9890

OP - 質問間の比較 (ANOVA):
F統計量: 0.26, p値: 0.9354

CO - 質問間の比較 (ANOVA):
F統計量: 0.09, p値: 0.9937

AG - 質問間の比較 (ANOVA):
F統計量: 0.29, p値: 0.9208

# ステップ5: セグメント分析
"""
参加者を以下の2つの距離に基づいて分類する：
a) Blended personaからの距離（average_distance）
b) Single Trait最大スコア（5）からの距離
各距離について、参加者を3つのグループ（近い、中間、遠い）に分類する。

各グループ内で、相対的選好スコアの平均を計算し、グループ間で比較する。
Tukey's Honest Significant Difference (HSD) テストを使用することができます。
これは、多重比較の問題を考慮しつつ、すべてのグループペア間の差の有意性を検定する方法です。

二元配置分散分析（Two-way ANOVA）を使用して、Blended距離とSingle Trait距離の両方が
相対的選好スコアに与える影響を分析する。
コメント: この修正により、Blended personaからの距離とSingle Trait最大スコアからの
距離の両方を考慮したセグメント分析が可能になります。
これにより、どのような特性プロファイルを持つ人々にBlended広告が特に効果的か、
あるいはSingleが効果的かをより詳細に理解できます。
"""

from statsmodels.stats.multicomp import pairwise_tukeyhsd
from statsmodels.formula.api import ols
from statsmodels.stats.anova import anova_lm

def segment_analysis(trait):
    # 距離のセグメント化
    data['blended_segment'] = pd.qcut(data['average_distance'], q=3, labels=['近い', '中間', '遠い'])
    data[f'{trait}_single_segment'] = pd.qcut(data[f'{trait}_distance_from_max'], q=3, labels=['近い', '中間', '遠い'])
    
    for question in questions:
        # グループごとの平均スコア計算
        grouped_means = data.groupby(['blended_segment', f'{trait}_single_segment'])[f'{trait}_{question}_relative_score'].mean().unstack()
        print(f"{trait} - {question} グループ平均:")
        print(grouped_means)
        print()
        
        # 二元配置分散分析
        model = ols(f'{trait}_{question}_relative_score ~ C(blended_segment) + C({trait}_single_segment) + C(blended_segment):C({trait}_single_segment)', data=data).fit()
        anova_table = anova_lm(model, typ=2)
        print(f"{trait} - {question} 二元配置分散分析結果:")
        print(anova_table)
        print()
        
        # Tukey's HSDテスト
        # blended_segmentに基づくグループ比較
        tukey_blended = pairwise_tukeyhsd(data[f'{trait}_{question}_relative_score'], data['blended_segment'])
        print("Tukey's HSD test for blended_segment:")
        print(tukey_blended)
        print()
        
        # single_segmentに基づくグループ比較
        tukey_single = pairwise_tukeyhsd(data[f'{trait}_{question}_relative_score'], data[f'{trait}_single_segment'])
        print(f"Tukey's HSD test for {trait}_single_segment:")
        print(tukey_single)
        print()
        
        # 交互作用のグループ比較
        data['interaction_group'] = data['blended_segment'].astype(str) + '_' + data[f'{trait}_single_segment'].astype(str)
        tukey_interaction = pairwise_tukeyhsd(data[f'{trait}_{question}_relative_score'], data['interaction_group'])
        print("Tukey's HSD test for interaction groups:")
        print(tukey_interaction)
        
        print("\n" + "="*50 + "\n")

for trait in traits:
    segment_analysis(trait)

EX - preference グループ平均:
EX_single_segment        近い         中間         遠い
blended_segment                                  
近い                -7.125000  -6.842105  25.400000
中間                 0.133333  11.875000  -4.800000
遠い                 3.300000  -4.000000  21.769231

EX - preference 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                         687.531217   2.0  0.315852   
C(EX_single_segment)                      4229.398706   2.0  1.942987   
C(blended_segment):C(EX_single_segment)   4849.562524   4.0  1.113945   
Residual                                 88158.417341  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.730061  
C(EX_single_segment)                     0.149895  
C(blended_segment):C(EX_single_segment)  0.355726  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -4.4821 0.8627 -25.1109 16.1466  False
    中間     遠い   6.9929 0.7065 -13.9543   27.94  False
    近い     遠い   11.475 0.3714  -8.7837 31.7337  False
-----------------------------------------------------

Tukey's HSD test for EX_single_segment:
Multiple Comparison of Means - Tukey HSD, FWER=0.05 
====================================================
group1 group2 meandiff p-adj  lower    upper  reject
----------------------------------------------------
    中間     近い   1.1863 0.988 -17.9358 20.3083  False
    中間     遠い  18.6355 0.095  -2.4909  39.762  False
    近い     遠い  17.4493 0.129   -3.806 38.7045  False
----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い -11.7417 0.9962 -57.7751 34.2917  False
 中間_中間  中間_遠い  -16.675 0.9931 -76.6184 43.2684  False
 中間_中間  近い_中間 -18.7171  0.914  -63.033 25.5988  False
 中間_中間  近い_近い    -19.0  0.964 -71.5738 33.5738  False
 中間_中間  近い_遠い   13.525 0.9984 -46.4184 73.4684  False
 中間_中間  遠い_中間  -15.875 0.9906 -70.2941 38.5441  False
 中間_中間  遠い_近い   -8.575 0.9998 -58.4509 41.3009  False
 中間_中間  遠い_遠い   9.8942 0.9991 -37.3548 57.1432  False
 中間_近い  中間_遠い  -4.9333    1.0 -59.2314 49.3647  False
 中間_近い  近い_中間  -6.9754 0.9995  -43.293 29.3421  False
 中間_近い  近い_近い  -7.2583 0.9999 -53.2917 38.7751  False
 中間_近い  近い_遠い  25.2667 0.8601 -29.0314 79.5647  False
 中間_近い  遠い_中間  -4.1333    1.0 -52.2634 43.9967  False
 中間_近い  遠い_近い   3.1667    1.0 -39.7597  46.093  False
 中間_近い  遠い_遠い  21.6359  0.726  -18.208 61.4798  False
 中間_遠い  近い_中間  -2.0421    1.0 -54.8919 50.8077  False
 中間_遠い  近い_近い   -2.325    1.0 -62.2684 57.6184  False
 中間_遠い  近い_遠い     30.2 0.8756 -36.3012 96.7012  False
 中間_遠い  遠い_中間      0.8    1.0 -60.7682 62.3682  False
 中間_遠い  遠い_近い      8.1    1.0 -49.4917 65.6917  False
 中間_遠い  遠い_遠い  26.5692 0.8379 -28.7631 81.9016  False
 近い_中間  近い_近い  -0.2829    1.0 -44.5988  44.033  False
 近い_中間  近い_遠い  32.2421 0.5857 -20.6077 85.0919  False
 近い_中間  遠い_中間   2.8421    1.0  -43.648 49.3322  False
 近い_中間  遠い_近い  10.1421 0.9969 -30.9371 51.2213  False
 近い_中間  遠い_遠い  28.6113 0.2932  -9.2352 66.4579  False
 近い_近い  近い_遠い   32.525 0.7268 -27.4184 92.4684  False
 近い_近い  遠い_中間    3.125    1.0 -51.2941 57.5441  False
 近い_近い  遠い_近い   10.425 0.9991 -39.4509 60.3009  False
 近い_近い  遠い_遠い  28.8942 0.5825 -18.3548 76.1432  False
 近い_遠い  遠い_中間    -29.4  0.842 -90.9682 32.1682  False
 近い_遠い  遠い_近い    -22.1  0.949 -79.6917 35.4917  False
 近い_遠い  遠い_遠い  -3.6308    1.0 -58.9631 51.7016  False
 遠い_中間  遠い_近い      7.3    1.0 -44.5173 59.1173  False
 遠い_中間  遠い_遠い  25.7692 0.7648 -23.5248 75.0632  False
 遠い_近い  遠い_遠い  18.4692 0.9189 -25.7582 62.6967  False
-----------------------------------------------------

==================================================

EX - believability グループ平均:
EX_single_segment        近い         中間         遠い
blended_segment                                  
近い                -4.875000  -7.894737  24.400000
中間                -3.666667  15.625000 -16.200000
遠い                -4.000000 -11.857143  23.153846

EX - believability 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                         168.177584   2.0  0.104446   
C(EX_single_segment)                      4940.012912   2.0  3.067982   
C(blended_segment):C(EX_single_segment)   9790.828818   4.0  3.040284   
Residual                                 65212.422258  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.900944  
C(EX_single_segment)                     0.051957  
C(blended_segment):C(EX_single_segment)  0.021780  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -1.7009 0.9744 -20.4054 17.0036  False
    中間     遠い   6.3262 0.7076 -12.6671 25.3194  False
    近い     遠い   8.0271 0.5526 -10.3419 26.3961  False
-----------------------------------------------------

Tukey's HSD test for EX_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -0.8841 0.9917 -18.0119 16.2437  False
    中間     遠い   18.046 0.0649  -0.8772 36.9692  False
    近い     遠い  18.9302 0.0517  -0.1084 37.9687  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い -19.2917 0.8266 -58.8836 20.3002  False
 中間_中間  中間_遠い  -31.825   0.57 -83.3805 19.7305  False
 中間_中間  近い_中間 -23.5197 0.5705 -61.6345  14.595  False
 中間_中間  近い_近い    -20.5 0.8767 -65.7171 24.7171  False
 中間_中間  近い_遠い    8.775 0.9998 -42.7805 60.3305  False
 中間_中間  遠い_中間 -27.4821 0.6349 -74.2863  19.322  False
 中間_中間  遠い_近い  -19.625 0.8711 -62.5217 23.2717  False
 中間_中間  遠い_遠い   7.5288 0.9996 -33.1086 48.1662  False
 中間_近い  中間_遠い -12.5333 0.9946 -59.2334 34.1667  False
 中間_近い  近い_中間  -4.2281    1.0 -35.4637 27.0075  False
 中間_近い  近い_近い  -1.2083    1.0 -40.8002 38.3836  False
 中間_近い  近い_遠い  28.0667 0.6053 -18.6334 74.7667  False
 中間_近い  遠い_中間  -8.1905 0.9994 -49.5856 33.2047  False
 中間_近い  遠い_近い  -0.3333    1.0  -37.253 36.5863  False
 中間_近い  遠い_遠い  26.8205 0.2502  -7.4479  61.089  False
 中間_遠い  近い_中間   8.3053 0.9997 -37.1492 53.7597  False
 中間_遠い  近い_近い   11.325 0.9987 -40.2305 62.8805  False
 中間_遠い  近い_遠い     40.6 0.3769 -16.5956 97.7956  False
 中間_遠い  遠い_中間   4.3429    1.0   -48.61 57.2957  False
 中間_遠い  遠い_近い     12.2  0.997 -37.3329 61.7329  False
 中間_遠い  遠い_遠い  39.3538 0.1885  -8.2358 86.9435  False
 近い_中間  近い_近い   3.0197    1.0  -35.095 41.1345  False
 近い_中間  近い_遠い  32.2947 0.3757 -13.1597 77.7492  False
 近い_中間  遠い_中間  -3.9624    1.0 -43.9471 36.0223  False
 近い_中間  遠い_近い   3.8947    1.0 -31.4362 39.2257  False
 近い_中間  遠い_遠い  31.0486 0.0735   -1.502 63.5992  False
 近い_近い  近い_遠い   29.275 0.6756 -22.2805 80.8305  False
 近い_近い  遠い_中間  -6.9821 0.9999 -53.7863  39.822  False
 近い_近い  遠い_近い    0.875    1.0 -42.0217 43.7717  False
 近い_近い  遠い_遠い  28.0288 0.4167 -12.6086 68.6662  False
 近い_遠い  遠い_中間 -36.2571 0.4269   -89.21 16.6957  False
 近い_遠い  遠い_近い    -28.4 0.6641 -77.9329 21.1329  False
 近い_遠い  遠い_遠い  -1.2462    1.0 -48.8358 46.3435  False
 遠い_中間  遠い_近い   7.8571 0.9997 -36.7093 52.4236  False
 遠い_中間  遠い_遠い   35.011 0.1899  -7.3852 77.4072  False
 遠い_近い  遠い_遠い  27.1538 0.3692 -10.8848 65.1925  False
-----------------------------------------------------

==================================================

EX - overall_liking グループ平均:
EX_single_segment         近い         中間         遠い
blended_segment                                   
近い                -12.750000  -3.684211  26.800000
中間                 -1.133333  10.000000 -20.200000
遠い                 14.500000 -13.857143  21.384615

EX - overall_liking 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                        1434.695240   2.0  0.662232   
C(EX_single_segment)                      2243.828843   2.0  1.035715   
C(blended_segment):C(EX_single_segment)  11517.568733   4.0  2.658162   
Residual                                 87741.372662  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.518468  
C(EX_single_segment)                     0.359625  
C(blended_segment):C(EX_single_segment)  0.038587  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   0.1696 0.9998 -20.9067  21.246  False
    中間     遠い  12.2238 0.3653  -9.1778 33.6255  False
    近い     遠い  12.0542 0.3512  -8.6441 32.7524  False
-----------------------------------------------------

Tukey's HSD test for EX_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   3.3467 0.9146 -16.4766   23.17  False
    中間     遠い  16.0806 0.1924  -5.8207 37.9818  False
    近い     遠い  12.7339 0.3568  -9.3009 34.7686  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
  Multiple Comparison of Means - Tukey HSD, FWER=0.05  
=======================================================
group1 group2 meandiff p-adj    lower    upper   reject
-------------------------------------------------------
 中間_中間  中間_近い -11.1333 0.9973  -57.0577   34.791  False
 中間_中間  中間_遠い    -30.2 0.7968  -90.0014  29.6014  False
 中間_中間  近い_中間 -13.6842 0.9861  -57.8952  30.5268  False
 中間_中間  近い_近い   -22.75 0.9012  -75.1993  29.6993  False
 中間_中間  近い_遠い     16.8 0.9926  -43.0014  76.6014  False
 中間_中間  遠い_中間 -23.8571 0.8944  -78.1474  30.4331  False
 中間_中間  遠い_近い      4.5    1.0  -45.2578  54.2578  False
 中間_中間  遠い_遠い  11.3846 0.9974  -35.7525  58.5217  False
 中間_近い  中間_遠い -19.0667 0.9692  -73.2361  35.1028  False
 中間_近い  近い_中間  -2.5509    1.0  -38.7824  33.6807  False
 中間_近い  近い_近い -11.6167 0.9964   -57.541  34.3077  False
 中間_近い  近い_遠い  27.9333 0.7779  -26.2361  82.1028  False
 中間_近い  遠い_中間 -12.7238  0.995  -60.7399  35.2923  False
 中間_近い  遠い_近い  15.6333 0.9618  -27.1914   58.458  False
 中間_近い  遠い_遠い  22.5179 0.6784  -17.2316  62.2674  False
 中間_遠い  近い_中間  16.5158  0.985  -36.2089  69.2404  False
 中間_遠い  近い_近い     7.45    1.0  -52.3514  67.2514  False
 中間_遠い  近い_遠い     47.0 0.3797  -19.3437 113.3437  False
 中間_遠い  遠い_中間   6.3429    1.0  -55.0795  67.7652  False
 中間_遠い  遠い_近い     34.7 0.5989  -22.7554  92.1554  False
 中間_遠い  遠い_遠い  41.5846 0.2977  -13.6167  96.7859  False
 近い_中間  近い_近い  -9.0658 0.9992  -53.2768  35.1452  False
 近い_中間  近い_遠い  30.4842  0.654  -22.2404  83.2089  False
 近い_中間  遠い_中間 -10.1729 0.9987  -56.5529  36.2071  False
 近い_中間  遠い_近い  18.1842 0.8892  -22.7977  59.1661  False
 近い_中間  遠い_遠い  25.0688 0.4699  -12.6881  62.8257  False
 近い_近い  近い_遠い    39.55 0.4754  -20.2514  99.3514  False
 近い_近い  遠い_中間  -1.1071    1.0  -55.3974  53.1831  False
 近い_近い  遠い_近い    27.25 0.7168  -22.5078  77.0078  False
 近い_近い  遠い_遠い  34.1346 0.3498  -13.0025  81.2717  False
 近い_遠い  遠い_中間 -40.6571 0.4742 -102.0795  20.7652  False
 近い_遠い  遠い_近い    -12.3 0.9989  -69.7554  45.1554  False
 近い_遠い  遠い_遠い  -5.4154    1.0  -60.6167  49.7859  False
 遠い_中間  遠い_近い  28.3571  0.715  -23.3375  80.0518  False
 遠い_中間  遠い_遠い  35.2418 0.3639  -13.9355   84.419  False
 遠い_近い  遠い_遠い   6.8846 0.9999  -37.2381  51.0073  False
-------------------------------------------------------

==================================================

EX - persuasiveness グループ平均:
EX_single_segment        近い        中間         遠い
blended_segment                                 
近い                -7.250000 -9.157895  24.600000
中間                 0.333333  6.125000 -23.200000
遠い                -3.000000 -5.000000  22.461538

EX - persuasiveness 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                         743.896734   2.0  0.388719   
C(EX_single_segment)                      3366.395189   2.0  1.759089   
C(blended_segment):C(EX_single_segment)   9263.939095   4.0  2.420407   
Residual                                 77505.465418  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.679183  
C(EX_single_segment)                     0.178716  
C(blended_segment):C(EX_single_segment)  0.054998  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   -1.192 0.9888 -21.0531 18.6692  False
    中間     遠い    9.781 0.4823 -10.3868 29.9487  False
    近い     遠い  10.9729 0.3763  -8.5319 30.4778  False
-----------------------------------------------------

Tukey's HSD test for EX_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   2.1907 0.9569 -16.2897 20.6712  False
    中間     遠い  17.7059 0.1026  -2.7117 38.1235  False
    近い     遠い  15.5152 0.1753  -5.0269 36.0572  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05  
======================================================
group1 group2 meandiff p-adj   lower    upper   reject
------------------------------------------------------
 中間_中間  中間_近い  -5.7917    1.0 -48.9542  37.3709  False
 中間_中間  中間_遠い  -29.325 0.7667 -85.5301  26.8801  False
 中間_中間  近い_中間 -15.2829 0.9601 -56.8351  26.2693  False
 中間_中間  近い_近い  -13.375 0.9942 -62.6701  35.9201  False
 中間_中間  近い_遠い   18.475 0.9797 -37.7301  74.6801  False
 中間_中間  遠い_中間  -11.125 0.9987 -62.1503  39.9003  False
 中間_中間  遠い_近い   -9.125 0.9994 -55.8905  37.6405  False
 中間_中間  遠い_遠い  16.3365 0.9595 -27.9658  60.6389  False
 中間_近い  中間_遠い -23.5333 0.8645 -74.4451  27.3784  False
 中間_近い  近い_中間  -9.4912  0.993 -43.5439  24.5614  False
 中間_近い  近い_近い  -7.5833 0.9997 -50.7459  35.5792  False
 中間_近い  近い_遠い  24.2667 0.8433 -26.6451  75.1784  False
 中間_近い  遠い_中間  -5.3333    1.0 -50.4618  39.7951  False
 中間_近い  遠い_近い  -3.3333    1.0 -43.5826   36.916  False
 中間_近い  遠い_遠い  22.1282 0.6239 -15.2308  59.4872  False
 中間_遠い  近い_中間  14.0421 0.9922 -35.5118   63.596  False
 中間_遠い  近い_近い    15.95 0.9921 -40.2551  72.1551  False
 中間_遠い  近い_遠い     47.8 0.2756 -14.5539 110.1539  False
 中間_遠い  遠い_中間     18.2 0.9844 -39.5285  75.9285  False
 中間_遠い  遠い_近い     20.2 0.9559 -33.8001  74.2001  False
 中間_遠い  遠い_遠い  45.6615 0.1299  -6.2201  97.5432  False
 近い_中間  近い_近い   1.9079    1.0 -39.6443  43.4601  False
 近い_中間  近い_遠い  33.7579  0.434  -15.796  83.3118  False
 近い_中間  遠い_中間   4.1579    1.0 -39.4329  47.7487  False
 近い_中間  遠い_近い   6.1579 0.9999 -32.3594  44.6752  False
 近い_中間  遠い_遠い  31.6194 0.1199  -3.8668  67.1057  False
 近い_近い  近い_遠い    31.85  0.678 -24.3551  88.0551  False
 近い_近い  遠い_中間     2.25    1.0 -48.7753  53.2753  False
 近い_近い  遠い_近い     4.25    1.0 -42.5155  51.0155  False
 近い_近い  遠い_遠い  29.7115 0.4559 -14.5908  74.0139  False
 近い_遠い  遠い_中間    -29.6 0.7832 -87.3285  28.1285  False
 近い_遠い  遠い_近い    -27.6 0.7861 -81.6001  26.4001  False
 近い_遠い  遠い_遠い  -2.1385    1.0 -54.0201  49.7432  False
 遠い_中間  遠い_近い      2.0    1.0 -46.5858  50.5858  False
 遠い_中間  遠い_遠い  27.4615   0.62 -18.7583  73.6814  False
 遠い_近い  遠い_遠い  25.4615 0.5772 -16.0077  66.9308  False
------------------------------------------------------

==================================================

EX - interest グループ平均:
EX_single_segment         近い         中間         遠い
blended_segment                                   
近い                -10.000000  -9.421053  25.600000
中間                  1.266667  12.375000  -8.400000
遠い                  6.300000  -2.714286  23.846154

EX - interest 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                        1783.649118   2.0  0.805568   
C(EX_single_segment)                      3970.355028   2.0  1.793174   
C(blended_segment):C(EX_single_segment)   6315.817216   4.0  1.426240   
Residual                                 89673.060791  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.450381  
C(EX_single_segment)                     0.172975  
C(blended_segment):C(EX_single_segment)  0.232763  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   -6.808 0.7186 -27.7235 14.1074  False
    中間     遠い   9.0857 0.5663 -12.1526  30.324  False
    近い     遠い  15.8938 0.1612  -4.6465  36.434  False
-----------------------------------------------------

Tukey's HSD test for EX_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   2.9724 0.9301 -16.5613  22.506  False
    中間     遠い  20.1292 0.0727  -1.4521 41.7104  False
    近い     遠い  17.1568 0.1494   -4.556 38.8696  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05  
======================================================
group1 group2 meandiff p-adj   lower    upper   reject
------------------------------------------------------
 中間_中間  中間_近い -11.1083 0.9975 -57.5355  35.3188  False
 中間_中間  中間_遠い  -20.775 0.9734 -81.2311  39.6811  False
 中間_中間  近い_中間 -21.7961  0.826  -66.491  22.8989  False
 中間_中間  近い_近い  -22.375 0.9144 -75.3985  30.6485  False
 中間_中間  近い_遠い   13.225 0.9987 -47.2311  73.6811  False
 中間_中間  遠い_中間 -15.0893 0.9936 -69.9739  39.7953  False
 中間_中間  遠い_近い   -6.075    1.0 -56.3775  44.2275  False
 中間_中間  遠い_遠い  11.4712 0.9974  -36.182  59.1243  False
 中間_近い  中間_遠い  -9.6667 0.9997 -64.4291  45.0958  False
 中間_近い  近い_中間 -10.6877 0.9905 -47.3159  25.9405  False
 中間_近い  近い_近い -11.2667 0.9973 -57.6938  35.1605  False
 中間_近い  近い_遠い  24.3333 0.8884 -30.4291  79.0958  False
 中間_近い  遠い_中間   -3.981    1.0 -52.5227  44.5608  False
 中間_近い  遠い_近い   5.0333    1.0 -38.2602  48.3269  False
 中間_近い  遠い_遠い  22.5795 0.6879 -17.6052  62.7642  False
 中間_遠い  近い_中間  -1.0211    1.0 -54.3229  52.2808  False
 中間_遠い  近い_近い     -1.6    1.0 -62.0561  58.8561  False
 中間_遠い  近い_遠い     34.0 0.7935 -33.0701 101.0701  False
 中間_遠い  遠い_中間   5.6857    1.0 -56.4091  67.7805  False
 中間_遠い  遠い_近い     14.7 0.9964 -43.3844  72.7844  False
 中間_遠い  遠い_遠い  32.2462 0.6547 -23.5595  88.0518  False
 近い_中間  近い_近い  -0.5789    1.0 -45.2739   44.116  False
 近い_中間  近い_遠い  35.0211 0.4846 -18.2808  88.3229  False
 近い_中間  遠い_中間   6.7068 0.9999  -40.181  53.5945  False
 近い_中間  遠い_近い  15.7211 0.9522 -25.7095  57.1516  False
 近い_中間  遠い_遠い  33.2672 0.1383  -4.9031  71.4375  False
 近い_近い  近い_遠い     35.6 0.6313 -24.8561  96.0561  False
 近い_近い  遠い_中間   7.2857    1.0 -47.5989  62.1703  False
 近い_近い  遠い_近い     16.3 0.9815 -34.0025  66.6025  False
 近い_近い  遠い_遠い  33.8462 0.3761  -13.807  81.4993  False
 近い_遠い  遠い_中間 -28.3143 0.8731 -90.4091  33.7805  False
 近い_遠い  遠い_近い    -19.3 0.9783 -77.3844  38.7844  False
 近い_遠い  遠い_遠い  -1.7538    1.0 -57.5595  54.0518  False
 遠い_中間  遠い_近い   9.0143 0.9998 -43.2463  61.2749  False
 遠い_中間  遠い_遠い  26.5604 0.7431 -23.1552  76.2761  False
 遠い_近い  遠い_遠い  17.5462 0.9414 -27.0596  62.1519  False
------------------------------------------------------

==================================================

EX - click_likelihood グループ平均:
EX_single_segment        近い         中間         遠い
blended_segment                                  
近い                -8.750000  -6.421053  25.600000
中間                 0.866667  12.250000  -6.400000
遠い                 1.200000  -6.428571  22.384615

EX - click_likelihood 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                         553.740073   2.0  0.228650   
C(EX_single_segment)                      4697.135916   2.0  1.939537   
C(blended_segment):C(EX_single_segment)   5640.015105   4.0  1.164435   
Residual                                 98082.156121  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.796119  
C(EX_single_segment)                     0.150389  
C(blended_segment):C(EX_single_segment)  0.332697  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -4.8214 0.8579  -26.604 16.9611  False
    中間     遠い   5.7786  0.808 -16.3402 27.8974  False
    近い     遠い     10.6 0.4673 -10.7918 31.9918  False
-----------------------------------------------------

Tukey's HSD test for EX_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   0.6658 0.9966 -19.5071 20.8386  False
    中間     遠い  18.8555 0.1141  -3.4319 41.1429  False
    近い     遠い  18.1897 0.1352  -4.2336  40.613  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05  
======================================================
group1 group2 meandiff p-adj   lower    upper   reject
------------------------------------------------------
 中間_中間  中間_近い -11.3833 0.9979 -59.9386  37.1719  False
 中間_中間  中間_遠い   -18.65 0.9898 -81.8773  44.5773  False
 中間_中間  近い_中間 -18.6711 0.9363 -65.4147  28.0726  False
 中間_中間  近い_近い    -21.0 0.9527  -76.454   34.454  False
 中間_中間  近い_遠い    13.35  0.999 -49.8773  76.5773  False
 中間_中間  遠い_中間 -18.6786  0.981 -76.0789  38.7218  False
 中間_中間  遠い_近い   -11.05  0.999 -63.6583  41.5583  False
 中間_中間  遠い_遠い  10.1346 0.9992 -39.7028  59.9721  False
 中間_近い  中間_遠い  -7.2667    1.0 -64.5393   50.006  False
 中間_近い  近い_中間  -7.2877 0.9995 -45.5949  31.0194  False
 中間_近い  近い_近い  -9.6167 0.9994 -58.1719  38.9386  False
 中間_近い  近い_遠い  24.7333 0.9034 -32.5393   82.006  False
 中間_近い  遠い_中間  -7.2952 0.9999  -58.062  43.4715  False
 中間_近い  遠い_近い   0.3333    1.0 -44.9447  45.6113  False
 中間_近い  遠い_遠い  21.5179 0.7845 -20.5087  63.5446  False
 中間_遠い  近い_中間  -0.0211    1.0 -55.7661   55.724  False
 中間_遠い  近い_近い    -2.35    1.0 -65.5773  60.8773  False
 中間_遠い  近い_遠い     32.0 0.8728 -38.1444 102.1444  False
 中間_遠い  遠い_中間  -0.0286    1.0 -64.9696  64.9125  False
 中間_遠い  遠い_近い      7.6    1.0 -53.1468  68.3468  False
 中間_遠い  遠い_遠い  28.7846 0.8169  -29.579  87.1482  False
 近い_中間  近い_近い  -2.3289    1.0 -49.0726  44.4147  False
 近い_中間  近い_遠い  32.0211 0.6619  -23.724  87.7661  False
 近い_中間  遠い_中間  -0.0075    1.0 -49.0445  49.0294  False
 近い_中間  遠い_近い   7.6211 0.9997 -35.7086  50.9507  False
 近い_中間  遠い_遠い  28.8057 0.3546 -11.1142  68.7256  False
 近い_近い  近い_遠い    34.35 0.7255 -28.8773  97.5773  False
 近い_近い  遠い_中間   2.3214    1.0 -55.0789  59.7218  False
 近い_近い  遠い_近い     9.95 0.9995 -42.6583  62.5583  False
 近い_近い  遠い_遠い  31.1346  0.554 -18.7028  80.9721  False
 近い_遠い  遠い_中間 -32.0286 0.8169 -96.9696  32.9125  False
 近い_遠い  遠い_近い    -24.4 0.9343 -85.1468  36.3468  False
 近い_遠い  遠い_遠い  -3.2154    1.0  -61.579  55.1482  False
 遠い_中間  遠い_近い   7.6286    1.0 -47.0275  62.2846  False
 遠い_中間  遠い_遠い  28.8132 0.7037 -23.1813  80.8076  False
 遠い_近い  遠い_遠い  21.1846 0.8757 -25.4657   67.835  False
------------------------------------------------------

==================================================

OP - preference グループ平均:
OP_single_segment         近い        中間         遠い
blended_segment                                  
近い                  0.615385 -7.562500 -20.333333
中間                 -8.583333 -6.833333 -12.700000
遠い                -23.200000 -0.666667   7.882353

OP - preference 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                         545.177888   2.0  0.266845   
C(OP_single_segment)                       870.980983   2.0  0.426314   
C(blended_segment):C(OP_single_segment)   6604.105364   4.0  1.616236   
Residual                                 82743.562462  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.766462  
C(OP_single_segment)                     0.654368  
C(blended_segment):C(OP_single_segment)  0.178125  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   4.2411 0.8672 -15.6292 24.1113  False
    中間     遠い   6.3452 0.7345 -13.8318 26.5222  False
    近い     遠い   2.1042 0.9642 -17.4097  21.618  False
-----------------------------------------------------

Tukey's HSD test for OP_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -2.7829 0.9416 -22.8538 17.2881  False
    中間     遠い     4.76 0.8484 -15.9962 25.5162  False
    近い     遠い   7.5429 0.6146 -11.5275 26.6132  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い    -1.75    1.0 -52.6837 49.1837  False
 中間_中間  中間_遠い  -5.8667    1.0 -58.4707 46.7374  False
 中間_中間  近い_中間  -0.7292    1.0 -49.4944 48.0361  False
 中間_中間  近い_近い   7.4487 0.9999 -42.8277 57.7251  False
 中間_中間  近い_遠い    -13.5 0.9996 -85.5311 58.5311  False
 中間_中間  遠い_中間   6.1667    1.0 -65.8644 78.1977  False
 中間_中間  遠い_近い -16.3667 0.9857 -68.9707 36.2374  False
 中間_中間  遠い_遠い  14.7157 0.9876 -33.6568 63.0882  False
 中間_近い  中間_遠い  -4.1167    1.0 -47.7336 39.5003  False
 中間_近い  近い_中間   1.0208    1.0 -37.8804 39.9221  False
 中間_近い  近い_近い   9.1987 0.9984 -31.5808 49.9783  False
 中間_近い  近い_遠い   -11.75 0.9997 -77.5051 54.0051  False
 中間_近い  遠い_中間   7.9167    1.0 -57.8384 73.6717  False
 中間_近い  遠い_近い -14.6167 0.9771 -58.2336 29.0003  False
 中間_近い  遠い_遠い  16.4657  0.907 -21.9421 54.8734  False
 中間_遠い  近い_中間   5.1375    1.0 -35.9265 46.2015  False
 中間_遠い  近い_近い  13.3154 0.9858 -29.5323  56.163  False
 中間_遠い  近い_遠い  -7.6333    1.0 -74.6906 59.4239  False
 中間_遠い  遠い_中間  12.0333 0.9997 -55.0239 79.0906  False
 中間_遠い  遠い_近い    -10.5 0.9981 -56.0564 35.0564  False
 中間_遠い  遠い_遠い  20.5824 0.7934 -20.0145 61.1792  False
 近い_中間  近い_近い   8.1779 0.9989 -29.8588 46.2145  False
 近い_中間  近い_遠い -12.7708 0.9993 -76.8609 51.3193  False
 近い_中間  遠い_中間   6.8958    1.0 -57.1943 70.9859  False
 近い_中間  遠い_近い -15.6375 0.9512 -56.7015 25.4265  False
 近い_中間  遠い_遠い  15.4449 0.8994 -20.0371 50.9268  False
 近い_近い  近い_遠い -20.9487 0.9825  -86.196 44.2986  False
 近い_近い  遠い_中間  -1.2821    1.0 -66.5293 63.9652  False
 近い_近い  遠い_近い -23.8154 0.7003  -66.663 19.0323  False
 近い_近い  遠い_遠い    7.267 0.9995 -30.2648 44.7988  False
 近い_遠い  遠い_中間  19.6667 0.9977 -63.5076 102.841  False
 近い_遠い  遠い_近い  -2.8667    1.0 -69.9239 64.1906  False
 近い_遠い  遠い_遠い  28.2157 0.8909 -35.5761 92.0075  False
 遠い_中間  遠い_近い -22.5333 0.9768 -89.5906 44.5239  False
 遠い_中間  遠い_遠い    8.549    1.0 -55.2428 72.3408  False
 遠い_近い  遠い_遠い  31.0824 0.2771  -9.5145 71.6792  False
-----------------------------------------------------

==================================================

OP - believability グループ平均:
OP_single_segment         近い         中間         遠い
blended_segment                                   
近い                  3.384615  -4.250000 -26.333333
中間                 -5.083333   1.333333  -8.800000
遠い                -20.300000 -13.000000   6.294118

OP - believability 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                         140.803890   2.0  0.103415   
C(OP_single_segment)                       386.634069   2.0  0.283969   
C(blended_segment):C(OP_single_segment)   6879.075965   4.0  2.526218   
Residual                                 55142.223002  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.901871  
C(OP_single_segment)                     0.753536  
C(blended_segment):C(OP_single_segment)  0.046983  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い    1.817 0.9628 -14.7093 18.3433  False
    中間     遠い   0.5357 0.9968 -16.2457 17.3171  False
    近い     遠い  -1.2812 0.9807 -17.5111 14.9486  False
-----------------------------------------------------

Tukey's HSD test for OP_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -2.3257  0.941 -19.0162 14.3648  False
    中間     遠い     1.96 0.9604 -15.3003 19.2203  False
    近い     遠い   4.2857  0.796 -11.5727 20.1441  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い  -6.4167 0.9999 -47.9962 35.1629  False
 中間_中間  中間_遠い -10.1333 0.9978 -53.0765 32.8099  False
 中間_中間  近い_中間  -5.5833    1.0 -45.3927 34.2261  False
 中間_中間  近い_近い   2.0513    1.0 -38.9917 43.0943  False
 中間_中間  近い_遠い -27.6667 0.8525  -86.469 31.1357  False
 中間_中間  遠い_中間 -14.3333 0.9972 -73.1357  44.469  False
 中間_中間  遠い_近い -21.6333  0.799 -64.5765 21.3099  False
 中間_中間  遠い_遠い   4.9608    1.0  -34.528 44.4496  False
 中間_近い  中間_遠い  -3.7167    1.0 -39.3233 31.8899  False
 中間_近い  近い_中間   0.8333    1.0 -30.9236 32.5902  False
 中間_近い  近い_近い   8.4679 0.9962 -24.8223 41.7582  False
 中間_近い  近い_遠い   -21.25 0.9393  -74.929  32.429  False
 中間_近い  遠い_中間  -7.9167 0.9999 -61.5956 45.7623  False
 中間_近い  遠い_近い -15.2167 0.9085 -50.8233 20.3899  False
 中間_近い  遠い_遠い  11.3775 0.9631 -19.9766 42.7315  False
 中間_遠い  近い_中間     4.55    1.0 -28.9725 38.0725  False
 中間_遠い  近い_近い  12.1846 0.9711  -22.794 47.1632  False
 中間_遠い  近い_遠い -17.5333 0.9828 -72.2754 37.2087  False
 中間_遠い  遠い_中間     -4.2    1.0  -58.942  50.542  False
 中間_遠い  遠い_近い    -11.5 0.9862 -48.6899 25.6899  False
 中間_遠い  遠い_遠い  15.0941 0.8739  -18.047 48.2352  False
 近い_中間  近い_近い   7.6346  0.997 -23.4165 38.6857  False
 近い_中間  近い_遠い -22.0833 0.9142 -74.4031 30.2365  False
 近い_中間  遠い_中間    -8.75 0.9998 -61.0698 43.5698  False
 近い_中間  遠い_近い   -16.05   0.84 -49.5725 17.4725  False
 近い_中間  遠い_遠い  10.5441 0.9624 -18.4215 39.5097  False
 近い_近い  近い_遠い -29.7179  0.696 -82.9824 23.5465  False
 近い_近い  遠い_中間 -16.3846 0.9867 -69.6491 36.8798  False
 近い_近い  遠い_近い -23.6846 0.4425 -58.6632  11.294  False
 近い_近い  遠い_遠い   2.9095    1.0 -27.7295 33.5485  False
 近い_遠い  遠い_中間  13.3333 0.9994 -54.5658 81.2325  False
 近い_遠い  遠い_近い   6.0333    1.0 -48.7087 60.7754  False
 近い_遠い  遠い_遠い  32.6275   0.55 -19.4488 84.7037  False
 遠い_中間  遠い_近い     -7.3    1.0  -62.042  47.442  False
 遠い_中間  遠い_遠い  19.2941 0.9583 -32.7821 71.3704  False
 遠い_近い  遠い_遠い  26.5941 0.2211   -6.547 59.7352  False
-----------------------------------------------------

==================================================

OP - overall_liking グループ平均:
OP_single_segment         近い        中間         遠い
blended_segment                                  
近い                  1.692308 -4.562500   9.333333
中間                -18.750000 -0.666667 -17.900000
遠い                -18.500000  3.666667  -0.941176

OP - overall_liking 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                        2708.926089   2.0  1.259604   
C(OP_single_segment)                       949.049125   2.0  0.441292   
C(blended_segment):C(OP_single_segment)   3420.978861   4.0  0.795348   
Residual                                 87099.964574  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.289265  
C(OP_single_segment)                     0.644742  
C(blended_segment):C(OP_single_segment)  0.531632  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  13.8527 0.2301  -6.1549 33.8603  False
    中間     遠い   8.2381 0.5997 -12.0783 28.5545  False
    近い     遠い  -5.6146  0.775 -25.2633 14.0341  False
-----------------------------------------------------

Tukey's HSD test for OP_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -8.4457 0.5881 -28.8858 11.9944  False
    中間     遠い  -2.9267 0.9417 -24.0646 18.2113  False
    近い     遠い    5.519 0.7771  -13.902 24.9401  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い -18.0833 0.9722 -70.3406 34.1739  False
 中間_中間  中間_遠い -17.2333 0.9831 -71.2044 36.7377  False
 中間_中間  近い_中間  -3.8958    1.0 -53.9284 46.1367  False
 中間_中間  近い_近い    2.359    1.0  -49.224 53.9419  False
 中間_中間  近い_遠い     10.0    1.0 -63.9029 83.9029  False
 中間_中間  遠い_中間   4.3333    1.0 -69.5696 78.2363  False
 中間_中間  遠い_近い -17.8333 0.9791 -71.8044 36.1377  False
 中間_中間  遠い_遠い  -0.2745    1.0 -49.9041 49.3551  False
 中間_近い  中間_遠い     0.85    1.0 -43.9005 45.6005  False
 中間_近い  近い_中間  14.1875 0.9673 -25.7247 54.0997  False
 中間_近い  近い_近い  20.4423 0.8245  -21.397 62.2816  False
 中間_近い  近い_遠い  28.0833 0.9203 -39.3805 95.5472  False
 中間_近い  遠い_中間  22.4167 0.9783 -45.0472 89.8805  False
 中間_近い  遠い_近い     0.25    1.0 -44.5005 45.0005  False
 中間_近い  遠い_遠い  17.8088 0.8786  -21.597 57.2147  False
 中間_遠い  近い_中間  13.3375  0.984 -28.7937 55.4687  False
 中間_遠い  近い_近い  19.5923 0.8867 -24.3688 63.5535  False
 中間_遠い  近い_遠い  27.2333 0.9394 -41.5666 96.0332  False
 中間_遠い  遠い_中間  21.5667  0.985 -47.2332 90.3666  False
 中間_遠い  遠い_近い     -0.6    1.0 -47.3403 46.1403  False
 中間_遠い  遠い_遠い  16.9588 0.9292  -24.693 58.6107  False
 近い_中間  近い_近い   6.2548 0.9999 -32.7703 45.2799  False
 近い_中間  近い_遠い  13.8958  0.999 -51.8598 79.6515  False
 近い_中間  遠い_中間   8.2292    1.0 -57.5265 73.9848  False
 近い_中間  遠い_近い -13.9375 0.9789 -56.0687 28.1937  False
 近い_中間  遠い_遠い   3.6213    1.0 -32.7827 40.0253  False
 近い_近い  近い_遠い    7.641    1.0 -59.3019 74.5839  False
 近い_近い  遠い_中間   1.9744    1.0 -64.9685 68.9172  False
 近い_近い  遠い_近い -20.1923 0.8686 -64.1535 23.7688  False
 近い_近い  遠い_遠い  -2.6335    1.0 -41.1406 35.8737  False
 近い_遠い  遠い_中間  -5.6667    1.0 -91.0024 79.6691  False
 近い_遠い  遠い_近い -27.8333 0.9317 -96.6332 40.9666  False
 近い_遠い  遠い_遠い -10.2745 0.9999 -75.7241  55.175  False
 遠い_中間  遠い_近い -22.1667 0.9821 -90.9666 46.6332  False
 遠い_中間  遠い_遠い  -4.6078    1.0 -70.0574 60.8417  False
 遠い_近い  遠い_遠い  17.5588 0.9148  -24.093 59.2107  False
-----------------------------------------------------

==================================================

OP - persuasiveness グループ平均:
OP_single_segment         近い         中間         遠い
blended_segment                                   
近い                 -0.923077 -10.125000 -36.333333
中間                -15.750000  -0.666667 -16.100000
遠い                -21.300000   1.333333   4.647059

OP - persuasiveness 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                         922.004742   2.0  0.529824   
C(OP_single_segment)                       543.140635   2.0  0.312112   
C(blended_segment):C(OP_single_segment)   8007.701257   4.0  2.300787   
Residual                                 70478.472097  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.590735  
C(OP_single_segment)                     0.732776  
C(blended_segment):C(OP_single_segment)  0.065684  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   3.7991 0.8776 -14.7982 22.3964  False
    中間     遠い   8.3095 0.5481 -10.5748 27.1939  False
    近い     遠い   4.5104 0.8264 -13.7533 22.7741  False
-----------------------------------------------------

Tukey's HSD test for OP_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -5.3486  0.778 -24.2128 13.5156  False
    中間     遠い   0.1133 0.9999 -19.3949 19.6215  False
    近い     遠い   5.4619 0.7484 -12.4618 23.3856  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
  Multiple Comparison of Means - Tukey HSD, FWER=0.05  
=======================================================
group1 group2 meandiff p-adj    lower    upper   reject
-------------------------------------------------------
 中間_中間  中間_近い -15.0833 0.9826  -62.0907   31.924  False
 中間_中間  中間_遠い -15.4333 0.9836  -63.9823  33.1157  False
 中間_中間  近い_中間  -9.4583  0.999  -54.4645  35.5478  False
 中間_中間  近い_近い  -0.2564    1.0  -46.6572  46.1444  False
 中間_中間  近い_遠い -35.6667 0.7387 -102.1451  30.8118  False
 中間_中間  遠い_中間      2.0    1.0  -64.4785  68.4785  False
 中間_中間  遠い_近い -20.6333 0.9111  -69.1823  27.9157  False
 中間_中間  遠い_遠い   5.3137    1.0  -39.3299  49.9574  False
 中間_近い  中間_遠い    -0.35    1.0  -40.6047  39.9047  False
 中間_近い  近い_中間    5.625 0.9999  -30.2775  41.5275  False
 中間_近い  近い_近い  14.8269 0.9409  -22.8091  52.4629  False
 中間_近い  近い_遠い -20.5833 0.9754  -81.2696  40.1029  False
 中間_近い  遠い_中間  17.0833 0.9925  -43.6029  77.7696  False
 中間_近い  遠い_近い    -5.55    1.0  -45.8047  34.7047  False
 中間_近い  遠い_遠い  20.3971 0.6598    -15.05  55.8441  False
 中間_遠い  近い_中間    5.975 0.9999  -31.9236  43.8736  False
 中間_遠い  近い_近い  15.1769  0.949  -24.3678  54.7216  False
 中間_遠い  近い_遠い -20.2333 0.9804  -82.1214  41.6547  False
 中間_遠い  遠い_中間  17.4333 0.9925  -44.4547  79.3214  False
 中間_遠い  遠い_近い     -5.2    1.0  -47.2447  36.8447  False
 中間_遠い  遠い_遠い  20.7471 0.7046  -16.7203  58.2144  False
 近い_中間  近い_近い   9.2019 0.9954  -25.9026  44.3065  False
 近い_中間  近い_遠い -26.2083 0.8899   -85.358  32.9413  False
 近い_中間  遠い_中間  11.4583 0.9995  -47.6913   70.608  False
 近い_中間  遠い_近い  -11.175 0.9899  -49.0736  26.7236  False
 近い_中間  遠い_遠い  14.7721 0.8797  -17.9747  47.5188  False
 近い_近い  近い_遠い -35.4103  0.633  -95.6279  24.8074  False
 近い_近い  遠い_中間   2.2564    1.0  -57.9612   62.474  False
 近い_近い  遠い_近い -20.3769 0.7786  -59.9216  19.1678  False
 近い_近い  遠い_遠い   5.5701 0.9999  -29.0685  40.2087  False
 近い_遠い  遠い_中間  37.6667 0.8211  -39.0961 114.4294  False
 近い_遠い  遠い_近い  15.0333 0.9973  -46.8547  76.9214  False
 近い_遠い  遠い_遠い  40.9804  0.404  -17.8939  99.8547  False
 遠い_中間  遠い_近い -22.6333 0.9614  -84.5214  39.2547  False
 遠い_中間  遠い_遠い   3.3137    1.0  -55.5606   62.188  False
 遠い_近い  遠い_遠い  25.9471 0.4111  -11.5203  63.4144  False
-------------------------------------------------------

==================================================

OP - interest グループ平均:
OP_single_segment         近い        中間         遠い
blended_segment                                  
近い                  1.076923 -9.437500 -14.666667
中間                -19.750000 -0.833333 -10.800000
遠い                -17.100000 -2.000000  -1.941176

OP - interest 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                         768.352015   2.0  0.389611   
C(OP_single_segment)                       525.478384   2.0  0.266456   
C(blended_segment):C(OP_single_segment)   3542.688613   4.0  0.898202   
Residual                                 79870.051753  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.678583  
C(OP_single_segment)                     0.766758  
C(blended_segment):C(OP_single_segment)  0.469056  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   6.8438 0.6722 -12.3224 26.0099  False
    中間     遠い      5.5 0.7793  -13.962  24.962  False
    近い     遠い  -1.3438 0.9842 -20.1661 17.4786  False
-----------------------------------------------------

Tukey's HSD test for OP_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -4.7771 0.8277    -24.2 14.6457  False
    中間     遠い   0.3133 0.9992 -19.7726 20.3993  False
    近い     遠い   5.0905 0.7885 -13.3641  23.545  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い -18.9167 0.9532 -68.9581 31.1248  False
 中間_中間  中間_遠い  -9.9667 0.9995 -61.6492 41.7159  False
 中間_中間  近い_中間  -8.6042 0.9997 -56.5152 39.3069  False
 中間_中間  近い_近い   1.9103    1.0 -47.4855  51.306  False
 中間_中間  近い_遠い -13.8333 0.9994 -84.6026 56.9359  False
 中間_中間  遠い_中間  -1.1667    1.0 -71.9359 69.6026  False
 中間_中間  遠い_近い -16.2667 0.9846 -67.9492 35.4159  False
 中間_中間  遠い_遠い  -1.1078    1.0  -48.633 46.4173  False
 中間_近い  中間_遠い     8.95 0.9991 -33.9029 51.8029  False
 中間_近い  近い_中間  10.3125 0.9944 -27.9073 48.5323  False
 中間_近い  近い_近い  20.8269 0.7703 -19.2383 60.8921  False
 中間_近い  近い_遠い   5.0833    1.0 -59.5199 69.6865  False
 中間_近い  遠い_中間    17.75 0.9937 -46.8532 82.3532  False
 中間_近い  遠い_近い     2.65    1.0 -40.2029 45.5029  False
 中間_近い  遠い_遠い  17.8088 0.8504 -19.9261 55.5438  False
 中間_遠い  近い_中間   1.3625    1.0 -38.9822 41.7072  False
 中間_遠い  近い_近い  11.8769 0.9924 -30.2202  53.974  False
 中間_遠い  近い_遠い  -3.8667    1.0 -69.7493 62.0159  False
 中間_遠い  遠い_中間      8.8    1.0 -57.0826 74.6826  False
 中間_遠い  遠い_近い     -6.3    1.0 -51.0584 38.4584  False
 中間_遠い  遠い_遠い   8.8588 0.9986 -31.0269 48.7445  False
 近い_中間  近い_近い  10.5144 0.9926 -26.8559 47.8848  False
 近い_中間  近い_遠い  -5.2292    1.0 -68.1966 57.7383  False
 近い_中間  遠い_中間   7.4375    1.0 -55.5299 70.4049  False
 近い_中間  遠い_近い  -7.6625 0.9995 -48.0072 32.6822  False
 近い_中間  遠い_遠い   7.4963 0.9989 -27.3641 42.3567  False
 近い_近い  近い_遠い -15.7436 0.9971 -79.8479 48.3607  False
 近い_近い  遠い_中間  -3.0769    1.0 -67.1813 61.0274  False
 近い_近い  遠い_近い -18.1769 0.9035  -60.274 23.9202  False
 近い_近い  遠い_遠い  -3.0181    1.0 -39.8924 33.8562  False
 近い_遠い  遠い_中間  12.6667 0.9999 -69.0506  94.384  False
 近い_遠い  遠い_近い  -2.4333    1.0 -68.3159 63.4493  False
 近い_遠い  遠い_遠い  12.7255 0.9992 -49.9488 75.3998  False
 遠い_中間  遠い_近い    -15.1 0.9982 -80.9826 50.7826  False
 遠い_中間  遠い_遠い   0.0588    1.0 -62.6155 62.7331  False
 遠い_近い  遠い_遠い  15.1588 0.9518 -24.7269 55.0445  False
-----------------------------------------------------

==================================================

OP - click_likelihood グループ平均:
OP_single_segment         近い        中間         遠い
blended_segment                                  
近い                  0.384615 -8.625000 -17.000000
中間                -21.250000 -0.666667 -11.700000
遠い                -19.500000  1.666667   1.529412

OP - click_likelihood 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                        1088.207212   2.0  0.510121   
C(OP_single_segment)                      1165.281498   2.0  0.546252   
C(blended_segment):C(OP_single_segment)   4531.130094   4.0  1.062034   
Residual                                 86395.912217  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.602339  
C(OP_single_segment)                     0.581235  
C(blended_segment):C(OP_single_segment)  0.380764  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   7.6786 0.6342  -12.397 27.7541  False
    中間     遠い   7.9619 0.6221 -12.4235 28.3473  False
    近い     遠い   0.2833 0.9994 -19.4321 19.9987  False
-----------------------------------------------------

Tukey's HSD test for OP_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -7.2343 0.6734 -27.5408 13.0722  False
    中間     遠い   0.7467  0.996 -20.2531 21.7464  False
    近い     遠い    7.981 0.5874 -11.3132 27.2751  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05  
======================================================
group1 group2 meandiff p-adj   lower    upper   reject
------------------------------------------------------
 中間_中間  中間_近い -20.5833 0.9397  -72.629  31.4623  False
 中間_中間  中間_遠い -11.0333 0.9992 -64.7858  42.7192  False
 中間_中間  近い_中間  -7.9583 0.9999 -57.7882  41.8716  False
 中間_中間  近い_近い   1.0513    1.0 -50.3228  52.4253  False
 中間_中間  近い_遠い -16.3333 0.9986  -89.937  57.2703  False
 中間_中間  遠い_中間   2.3333    1.0 -71.2703   75.937  False
 中間_中間  遠い_近い -18.8333   0.97 -72.5858  34.9192  False
 中間_中間  遠い_遠い   2.1961    1.0 -47.2325  51.6247  False
 中間_近い  中間_遠い     9.55 0.9989 -35.0192  54.1192  False
 中間_近い  近い_中間   12.625 0.9837 -27.1255  52.3755  False
 中間_近い  近い_近い  21.6346 0.7714 -20.0352  63.3045  False
 中間_近い  近い_遠い     4.25    1.0 -62.9406  71.4406  False
 中間_近い  遠い_中間  22.9167 0.9745  -44.274  90.1073  False
 中間_近い  遠い_近い     1.75    1.0 -42.8192  46.3192  False
 中間_近い  遠い_遠い  22.7794 0.6492 -16.4668  62.0257  False
 中間_遠い  近い_中間    3.075    1.0 -38.8855  45.0355  False
 中間_遠い  近い_近い  12.0846 0.9935 -31.6985  55.8677  False
 中間_遠い  近い_遠い     -5.3    1.0 -73.8213  63.2213  False
 中間_遠い  遠い_中間  13.3667 0.9994 -55.1546  81.8879  False
 中間_遠い  遠い_近い     -7.8 0.9998  -54.351   38.751  False
 中間_遠い  遠い_遠い  13.2294 0.9832 -28.2537  54.7126  False
 近い_中間  近い_近い   9.0096  0.998 -29.8574  47.8767  False
 近い_中間  近い_遠い   -8.375    1.0 -73.8643  57.1143  False
 近い_中間  遠い_中間  10.2917 0.9999 -55.1977   75.781  False
 近い_中間  遠い_近い  -10.875 0.9957 -52.8355  31.0855  False
 近い_中間  遠い_遠い  10.1544 0.9928 -26.1022   46.411  False
 近い_近い  近い_遠い -17.3846 0.9956 -84.0564  49.2872  False
 近い_近い  遠い_中間   1.2821    1.0 -65.3897  67.9538  False
 近い_近い  遠い_近い -19.8846 0.8756 -63.6677  23.8985  False
 近い_近い  遠い_遠い   1.1448    1.0 -37.2064   39.496  False
 近い_遠い  遠い_中間  18.6667 0.9987 -66.3235 103.6568  False
 近い_遠い  遠い_近い     -2.5    1.0 -71.0213  66.0213  False
 近い_遠い  遠い_遠い  18.5294  0.992 -46.6551  83.7139  False
 遠い_中間  遠い_近い -21.1667 0.9863 -89.6879  47.3546  False
 遠い_中間  遠い_遠い  -0.1373    1.0 -65.3217  65.0472  False
 遠い_近い  遠い_遠い  21.0294 0.7935 -20.4537  62.5126  False
------------------------------------------------------

==================================================

CO - preference グループ平均:
CO_single_segment        近い         中間         遠い
blended_segment                                  
近い                -2.500000   6.785714   4.250000
中間                -8.937500 -15.500000 -12.500000
遠い                 4.727273 -19.333333   8.076923

CO - preference 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                        3549.700477   2.0  1.835519   
C(CO_single_segment)                       716.571248   2.0  0.370533   
C(blended_segment):C(CO_single_segment)   3280.341119   4.0  0.848118   
Residual                                 78322.732871  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.166104  
C(CO_single_segment)                     0.691531  
C(blended_segment):C(CO_single_segment)  0.498858  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  14.3571 0.1744  -4.6233 33.3376  False
    中間     遠い  12.4738  0.276  -6.7997 31.7473  False
    近い     遠い  -1.8833 0.9685 -20.5233 16.7567  False
-----------------------------------------------------

Tukey's HSD test for CO_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   1.2495 0.9867  -17.841 20.3399  False
    中間     遠い    6.755 0.7127 -13.7426 27.2526  False
    近い     遠い   5.5055 0.7669 -13.3762 24.3872  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い   6.5625    1.0 -40.8822 54.0072  False
 中間_中間  中間_遠い      3.0    1.0 -54.2204 60.2204  False
 中間_中間  近い_中間  22.2857 0.8665 -26.0744 70.6458  False
 中間_中間  近い_近い     13.0 0.9963 -38.1795 64.1795  False
 中間_中間  近い_遠い    19.75 0.9593 -33.7748 73.2748  False
 中間_中間  遠い_中間  -3.8333    1.0 -61.0538 53.3871  False
 中間_中間  遠い_近い  20.2273 0.9339 -30.0723 70.5268  False
 中間_中間  遠い_遠い  23.5769  0.835  -25.338 72.4918  False
 中間_近い  中間_遠い  -3.5625    1.0 -51.0072 43.8822  False
 中間_近い  近い_中間  15.7232 0.9015 -20.5468 51.9933  False
 中間_近い  近い_近い   6.4375 0.9999 -33.5145 46.3895  False
 中間_近い  近い_遠い  13.1875 0.9868 -29.7278 56.1028  False
 中間_近い  遠い_中間 -10.3958 0.9987 -57.8405 37.0488  False
 中間_近い  遠い_近い  13.6648 0.9692 -25.1536 52.4831  False
 中間_近い  遠い_遠い  17.0144 0.8679 -19.9922  54.021  False
 中間_遠い  近い_中間  19.2857 0.9368 -29.0744 67.6458  False
 中間_遠い  近い_近い     10.0 0.9994 -41.1795 61.1795  False
 中間_遠い  近い_遠い    16.75 0.9851 -36.7748 70.2748  False
 中間_遠い  遠い_中間  -6.8333    1.0 -64.0538 50.3871  False
 中間_遠い  遠い_近い  17.2273 0.9739 -33.0723 67.5268  False
 中間_遠い  遠い_遠い  20.5769 0.9157  -28.338 69.4918  False
 近い_中間  近い_近い  -9.2857 0.9983 -50.3206 31.7492  False
 近い_中間  近い_遠い  -2.5357    1.0 -46.4609 41.3895  False
 近い_中間  遠い_中間  -26.119 0.7317 -74.4791  22.241  False
 近い_中間  遠い_近い  -2.0584    1.0 -41.9905 37.8736  False
 近い_中間  遠い_遠い   1.2912    1.0 -36.8819 39.4643  False
 近い_近い  近い_遠い     6.75 0.9999 -40.2614 53.7614  False
 近い_近い  遠い_中間 -16.8333 0.9796 -68.0128 34.3462  False
 近い_近い  遠い_近い   7.2273 0.9998 -36.0764  50.531  False
 近い_近い  遠い_遠い  10.5769 0.9963 -31.1104 52.2642  False
 近い_遠い  遠い_中間 -23.5833  0.893 -77.1081 29.9415  False
 近い_遠い  遠い_近い   0.4773    1.0 -45.5746 46.5291  False
 近い_遠い  遠い_遠い   3.8269    1.0 -40.7084 48.3622  False
 遠い_中間  遠い_近い  24.0606 0.8407 -26.2389 74.3602  False
 遠い_中間  遠い_遠い  27.4103  0.691 -21.5046 76.3252  False
 遠い_近い  遠い_遠い   3.3497    1.0 -37.2525 43.9518  False
-----------------------------------------------------

==================================================

CO - believability グループ平均:
CO_single_segment        近い         中間        遠い
blended_segment                                 
近い                 0.000000   6.000000  9.250000
中間                -6.375000 -12.166667 -6.666667
遠い                 7.636364 -17.833333  0.153846

CO - believability 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                        2793.648687   2.0  1.705229   
C(CO_single_segment)                       759.149125   2.0  0.463381   
C(blended_segment):C(CO_single_segment)   2338.645256   4.0  0.713749   
Residual                                 66350.487762  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.188189  
C(CO_single_segment)                     0.630810  
C(blended_segment):C(CO_single_segment)  0.584920  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  12.6161 0.2015  -4.8175 30.0496  False
    中間     遠い   6.9786 0.6166 -10.7241 24.6813  False
    近い     遠い  -5.6375 0.7131 -22.7583 11.4833  False
-----------------------------------------------------

Tukey's HSD test for CO_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   3.2058 0.9002 -14.2853 20.6969  False
    中間     遠い   5.0256 0.7996 -13.7547  23.806  False
    近い     遠い   1.8198 0.9659   -15.48 19.1197  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い   5.7917    1.0 -37.8766 49.4599  False
 中間_中間  中間_遠い      5.5    1.0 -47.1659 58.1659  False
 中間_中間  近い_中間  18.1667 0.9283 -26.3441 62.6774  False
 中間_中間  近い_近い  12.1667 0.9958 -34.9391 59.2724  False
 中間_中間  近い_遠い  21.4167 0.9001 -27.8477 70.6811  False
 中間_中間  遠い_中間  -5.6667    1.0 -58.3325 46.9992  False
 中間_中間  遠い_近い   19.803 0.9081 -26.4928 66.0989  False
 中間_中間  遠い_遠い  12.3205 0.9938 -32.7009 57.3419  False
 中間_近い  中間_遠い  -0.2917    1.0 -43.9599 43.3766  False
 中間_近い  近い_中間   12.375 0.9582 -21.0081 45.7581  False
 中間_近い  近い_近い    6.375 0.9998 -30.3969 43.1469  False
 中間_近い  近い_遠い   15.625 0.9396 -23.8744 55.1244  False
 中間_近い  遠い_中間 -11.4583 0.9954 -55.1266 32.2099  False
 中間_近い  遠い_近い  14.0114 0.9424 -21.7172 49.7399  False
 中間_近い  遠い_遠い   6.5288 0.9995 -27.5321 40.5898  False
 中間_遠い  近い_中間  12.6667  0.992 -31.8441 57.1774  False
 中間_遠い  近い_近い   6.6667 0.9999 -40.4391 53.7724  False
 中間_遠い  近い_遠い  15.9167 0.9818 -33.3477 65.1811  False
 中間_遠い  遠い_中間 -11.1667  0.999 -63.8325 41.4992  False
 中間_遠い  遠い_近い   14.303 0.9863 -31.9928 60.5989  False
 中間_遠い  遠い_遠い   6.8205 0.9999 -38.2009 51.8419  False
 近い_中間  近い_近い     -6.0 0.9999 -43.7686 31.7686  False
 近い_中間  近い_遠い     3.25    1.0 -37.1789 43.6789  False
 近い_中間  遠い_中間 -23.8333 0.7408 -68.3441 20.6774  False
 近い_中間  遠い_近い   1.6364    1.0 -35.1172 38.3899  False
 近い_中間  遠い_遠い  -5.8462 0.9998 -40.9808 29.2885  False
 近い_近い  近い_遠い     9.25 0.9989 -34.0194 52.5194  False
 近い_近い  遠い_中間 -17.8333 0.9528 -64.9391 29.2724  False
 近い_近い  遠い_近い   7.6364 0.9995 -32.2205 47.4932  False
 近い_近い  遠い_遠い   0.1538    1.0 -38.2153  38.523  False
 近い_遠い  遠い_中間 -27.0833 0.7126 -76.3477 22.1811  False
 近い_遠い  遠い_近い  -1.6136    1.0 -43.9999 40.7726  False
 近い_遠い  遠い_遠い  -9.0962 0.9986 -50.0866 31.8943  False
 遠い_中間  遠い_近い  25.4697 0.7118 -20.8262 71.7656  False
 遠い_中間  遠い_遠い  17.9872 0.9362 -27.0343 63.0086  False
 遠い_近い  遠い_遠い  -7.4825 0.9993 -44.8529 29.8879  False
-----------------------------------------------------

==================================================

CO - overall_liking グループ平均:
CO_single_segment        近い         中間    遠い
blended_segment                             
近い                -3.400000   1.857143  -1.0
中間                -9.187500 -17.000000 -12.0
遠い                 7.272727 -27.000000   7.0

CO - overall_liking 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                        2785.900299   2.0  1.342870   
C(CO_single_segment)                      1553.964153   2.0  0.749048   
C(blended_segment):C(CO_single_segment)   4468.566528   4.0  1.076978   
Residual                                 84020.733604  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.266845  
C(CO_single_segment)                     0.476063  
C(blended_segment):C(CO_single_segment)  0.373413  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  10.9643 0.3896  -8.8867 30.8152  False
    中間     遠い  11.7643 0.3497  -8.3931 31.9217  False
    近い     遠い      0.8 0.9947 -18.6948 20.2948  False
-----------------------------------------------------

Tukey's HSD test for CO_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   6.4241 0.7193 -13.3407 26.1889  False
    中間     遠い   9.5613 0.5325 -11.6604 30.7829  False
    近い     遠い   3.1371 0.9225 -16.4116 22.6858  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い   7.8125 0.9999 -41.3277 56.9527  False
 中間_中間  中間_遠い      5.0    1.0 -54.2653 64.2653  False
 中間_中間  近い_中間  18.8571 0.9543 -31.2312 68.9454  False
 中間_中間  近い_近い     13.6  0.996 -39.4085 66.6085  False
 中間_中間  近い_遠い     16.0 0.9912 -39.4376 71.4376  False
 中間_中間  遠い_中間    -10.0 0.9998 -69.2653 49.2653  False
 中間_中間  遠い_近い  24.2727 0.8592 -27.8244 76.3698  False
 中間_中間  遠い_遠い     24.0 0.8477  -26.663  74.663  False
 中間_近い  中間_遠い  -2.8125    1.0 -51.9527 46.3277  False
 中間_近い  近い_中間  11.0446 0.9901 -26.5216 48.6109  False
 中間_近い  近い_近い   5.7875    1.0 -35.5922 47.1672  False
 中間_近い  近い_遠い   8.1875 0.9996 -36.2615 52.6365  False
 中間_近い  遠い_中間 -17.8125 0.9633 -66.9527 31.3277  False
 中間_近い  遠い_近い  16.4602 0.9271 -23.7454 56.6658  False
 中間_近い  遠い_遠い  16.1875  0.914 -22.1416 54.5166  False
 中間_遠い  近い_中間  13.8571 0.9934 -36.2312 63.9454  False
 中間_遠い  近い_近い      8.6 0.9999 -44.4085 61.6085  False
 中間_遠い  近い_遠い     11.0 0.9994 -44.4376 66.4376  False
 中間_遠い  遠い_中間    -15.0 0.9964 -74.2653 44.2653  False
 中間_遠い  遠い_近い  19.2727 0.9587 -32.8244 71.3698  False
 中間_遠い  遠い_遠い     19.0 0.9553  -31.663  69.663  False
 近い_中間  近い_近い  -5.2571    1.0 -47.7585 37.2442  False
 近い_中間  近い_遠い  -2.8571    1.0 -48.3521 42.6378  False
 近い_中間  遠い_中間 -28.8571 0.6583 -78.9454 21.2312  False
 近い_中間  遠い_近い   5.4156    1.0 -35.9435 46.7746  False
 近い_中間  遠い_遠い   5.1429    1.0 -34.3945 44.6802  False
 近い_近い  近い_遠い      2.4    1.0 -46.2914 51.0914  False
 近い_近い  遠い_中間    -23.6 0.8873 -76.6085 29.4085  False
 近い_近い  遠い_近い  10.6727 0.9976 -34.1785 55.5239  False
 近い_近い  遠い_遠い     10.4 0.9974 -32.7771 53.5771  False
 近い_遠い  遠い_中間    -26.0 0.8548 -81.4376 29.4376  False
 近い_遠い  遠い_近い   8.2727 0.9998 -39.4249 55.9703  False
 近い_遠い  遠い_遠い      8.0 0.9998 -38.1269 54.1269  False
 遠い_中間  遠い_近い  34.2727 0.4828 -17.8244 86.3698  False
 遠い_中間  遠い_遠い     34.0  0.455  -16.663  84.663  False
 遠い_近い  遠い_遠い  -0.2727    1.0 -42.3259 41.7804  False
-----------------------------------------------------

==================================================

CO - persuasiveness グループ平均:
CO_single_segment         近い         中間        遠い
blended_segment                                  
近い                 -8.800000   4.357143  6.750000
中間                -11.500000 -18.666667 -3.833333
遠い                  9.363636 -19.000000  3.615385

CO - persuasiveness 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                        2800.405491   2.0  1.543961   
C(CO_single_segment)                      1144.470450   2.0  0.630986   
C(blended_segment):C(CO_single_segment)   4155.123542   4.0  1.145432   
Residual                                 73458.103330  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.219751  
C(CO_single_segment)                     0.534662  
C(blended_segment):C(CO_single_segment)  0.341213  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
Multiple Comparison of Means - Tukey HSD, FWER=0.05 
====================================================
group1 group2 meandiff p-adj   lower   upper  reject
----------------------------------------------------
    中間     近い  12.2366 0.2633 -6.3287 30.8019  False
    中間     遠い  12.5929 0.2542  -6.259 31.4447  False
    近い     遠い   0.3562 0.9988 -17.876 18.5885  False
----------------------------------------------------

Tukey's HSD test for CO_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   1.7786 0.9716  -16.773 20.3302  False
    中間     遠い    9.235 0.5133  -10.684 29.1541  False
    近い     遠い   7.4565 0.5983 -10.8923 25.8052  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い   7.1667 0.9999  -38.781 53.1143  False
 中間_中間  中間_遠い  14.8333 0.9947 -40.5816 70.2483  False
 中間_中間  近い_中間  23.0238 0.8196 -23.8104  69.858  False
 中間_中間  近い_近い   9.8667 0.9993  -39.698 59.4313  False
 中間_中間  近い_遠い  25.4167 0.8217 -26.4193 77.2526  False
 中間_中間  遠い_中間  -0.3333    1.0 -55.7483 55.0816  False
 中間_中間  遠い_近い  28.0303 0.6598 -20.6822 76.7428  False
 中間_中間  遠い_遠い  22.2821 0.8527 -25.0894 69.6535  False
 中間_近い  中間_遠い   7.6667 0.9998  -38.281 53.6143  False
 中間_近い  近い_中間  15.8571 0.8792 -19.2685 50.9828  False
 中間_近い  近い_近い      2.7    1.0 -35.9914 41.3914  False
 中間_近い  近い_遠い    18.25 0.8948 -23.3112 59.8112  False
 中間_近い  遠い_中間     -7.5 0.9998 -53.4477 38.4477  False
 中間_近い  遠い_近い  20.8636  0.702 -16.7299 58.4572  False
 中間_近い  遠い_遠い  15.1154 0.9146 -20.7235 50.9543  False
 中間_遠い  近い_中間   8.1905 0.9997 -38.6437 55.0247  False
 中間_遠い  近い_近い  -4.9667    1.0 -54.5313  44.598  False
 中間_遠い  近い_遠い  10.5833 0.9992 -41.2526 62.4193  False
 中間_遠い  遠い_中間 -15.1667 0.9938 -70.5816 40.2483  False
 中間_遠い  遠い_近い   13.197 0.9942 -35.5155 61.9094  False
 中間_遠い  遠い_遠い   7.4487 0.9999 -39.9228 54.8202  False
 近い_中間  近い_近い -13.1571 0.9788 -52.8973  26.583  False
 近い_中間  近い_遠い   2.3929    1.0 -40.1464 44.9321  False
 近い_中間  遠い_中間 -23.3571 0.8077 -70.1913  23.477  False
 近い_中間  遠い_近い   5.0065    1.0 -33.6656 43.6785  False
 近い_中間  遠い_遠い  -0.7418    1.0 -37.7104 36.2269  False
 近い_近い  近い_遠い    15.55 0.9743  -29.978  61.078  False
 近い_近い  遠い_中間    -10.2 0.9992 -59.7646 39.3646  False
 近い_近い  遠い_近い  18.1636  0.902 -23.7737  60.101  False
 近い_近い  遠い_遠い  12.4154 0.9867 -27.9566 52.7873  False
 近い_遠い  遠い_中間   -25.75  0.811 -77.5859 26.0859  False
 近い_遠い  遠い_近い   2.6136    1.0 -41.9852 47.2124  False
 近い_遠い  遠い_遠い  -3.1346    1.0 -46.2647 39.9955  False
 遠い_中間  遠い_近い  28.3636 0.6453 -20.3488 77.0761  False
 遠い_中間  遠い_遠い  22.6154 0.8421 -24.7561 69.9869  False
 遠い_近い  遠い_遠い  -5.7483 0.9999 -45.0693 33.5728  False
-----------------------------------------------------

==================================================

CO - interest グループ平均:
CO_single_segment        近い         中間         遠い
blended_segment                                  
近い                -6.500000   4.857143   9.250000
中間                -7.250000 -23.166667 -14.666667
遠い                 6.818182 -16.000000   7.153846

CO - interest 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                        4222.747678   2.0  2.136318   
C(CO_single_segment)                      1270.394098   2.0  0.642702   
C(blended_segment):C(CO_single_segment)   3671.565028   4.0  0.928736   
Residual                                 80054.209624  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.124695  
C(CO_single_segment)                     0.528530  
C(blended_segment):C(CO_single_segment)  0.451528  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  14.6562 0.1717  -4.6303 33.9428  False
    中間     遠い    14.65 0.1811  -4.9343 34.2343  False
    近い     遠い  -0.0063    1.0 -18.9469 18.9344  False
-----------------------------------------------------

Tukey's HSD test for CO_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   3.5582 0.9001 -15.8431 22.9595  False
    中間     遠い    9.349 0.5351 -11.4824 30.1804  False
    近い     遠い   5.7908 0.7526 -13.3983 24.9799  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い  15.9167 0.9785 -32.0496 63.8829  False
 中間_中間  中間_遠い      8.5 0.9999 -49.3494 66.3494  False
 中間_中間  近い_中間  28.0238 0.6645 -20.8679 76.9155  False
 中間_中間  近い_近い  16.6667 0.9821 -35.0755 68.4088  False
 中間_中間  近い_遠い  32.4167 0.6094 -21.6965 86.5299  False
 中間_中間  遠い_中間   7.1667    1.0 -50.6828 65.0161  False
 中間_中間  遠い_近い  29.9848 0.6296 -20.8677 80.8373  False
 中間_中間  遠い_遠い  30.3205 0.5791 -19.1321 79.7731  False
 中間_近い  中間_遠い  -7.4167 0.9999 -55.3829 40.5496  False
 中間_近い  近い_中間  12.1071 0.9792 -24.5616 48.7759  False
 中間_近い  近い_近い     0.75    1.0 -39.6412 41.1412  False
 中間_近い  近い_遠い     16.5 0.9516 -26.8871 59.8871  False
 中間_近い  遠い_中間    -8.75 0.9997 -56.7162 39.2162  False
 中間_近い  遠い_近い  14.0682 0.9657 -25.1769 53.3133  False
 中間_近い  遠い_遠い  14.4038 0.9481 -23.0096 51.8172  False
 中間_遠い  近い_中間  19.5238 0.9363 -29.3679 68.4155  False
 中間_遠い  近い_近い   8.1667 0.9999 -43.5755 59.9088  False
 中間_遠い  近い_遠い  23.9167 0.8913 -30.1965 78.0299  False
 中間_遠い  遠い_中間  -1.3333    1.0 -59.1828 56.5161  False
 中間_遠い  遠い_近い  21.4848 0.9138 -29.3677 72.3373  False
 中間_遠い  遠い_遠い  21.8205 0.8922 -27.6321 71.2731  False
 近い_中間  近い_近い -11.3571 0.9938 -52.8431 30.1288  False
 近い_中間  近い_遠い   4.3929    1.0 -40.0152  48.801  False
 近い_中間  遠い_中間 -20.8571 0.9093 -69.7488 28.0346  False
 近い_中間  遠い_近い    1.961    1.0   -38.41  42.332  False
 近い_中間  遠い_遠い   2.2967    1.0 -36.2961 40.8895  False
 近い_近い  近い_遠い    15.75 0.9787 -31.7782 63.2782  False
 近い_近い  遠い_中間     -9.5 0.9996 -61.2421 42.2421  False
 近い_近い  遠い_近い  13.3182 0.9876 -30.4616 57.0979  False
 近い_近い  遠い_遠い  13.6538 0.9815 -28.4917 55.7994  False
 近い_遠い  遠い_中間   -25.25 0.8582 -79.3632 28.8632  False
 近い_遠い  遠い_近い  -2.4318    1.0 -48.9899 44.1263  False
 近い_遠い  遠い_遠い  -2.0962    1.0 -47.1211 42.9288  False
 遠い_中間  遠い_近い  22.8182 0.8828 -28.0343 73.6707  False
 遠い_中間  遠い_遠い  23.1538 0.8559 -26.2988 72.6065  False
 遠い_近い  遠い_遠い   0.3357    1.0 -40.7128 41.3842  False
-----------------------------------------------------

==================================================

CO - click_likelihood グループ平均:
CO_single_segment         近い         中間         遠い
blended_segment                                   
近い                 -5.500000   6.571429   7.875000
中間                -13.000000 -22.500000 -12.166667
遠い                  7.454545 -17.000000   4.307692

CO - click_likelihood 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                        5581.678828   2.0  2.377340   
C(CO_single_segment)                       817.864049   2.0  0.348343   
C(blended_segment):C(CO_single_segment)   3265.231114   4.0  0.695361   
Residual                                 95088.633408  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.099243  
C(CO_single_segment)                     0.706909  
C(blended_segment):C(CO_single_segment)  0.597332  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  17.9821 0.1047  -2.8507  38.815  False
    中間     遠い  16.0571 0.1723  -5.0973 37.2116  False
    近い     遠い   -1.925 0.9726 -22.3842 18.5342  False
-----------------------------------------------------

Tukey's HSD test for CO_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い    0.685 0.9967  -20.406 21.7761  False
    中間     遠い   7.2806 0.7244 -15.3651 29.9263  False
    近い     遠い   6.5956 0.7321 -14.2648  27.456  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い      9.5 0.9997 -42.7767 61.7767  False
 中間_中間  中間_遠い  10.3333 0.9998 -52.7147 73.3813  False
 中間_中間  近い_中間  29.0714 0.7209 -24.2139 82.3567  False
 中間_中間  近い_近い     17.0 0.9883 -39.3919 73.3919  False
 中間_中間  近い_遠い   30.375  0.779  -28.601  89.351  False
 中間_中間  遠い_中間      5.5    1.0  -57.548  68.548  False
 中間_中間  遠い_近い  29.9545  0.731 -25.4678 85.3768  False
 中間_中間  遠い_遠い  26.8077 0.8099 -27.0889 80.7043  False
 中間_近い  中間_遠い   0.8333    1.0 -51.4433   53.11  False
 中間_近い  近い_中間  19.5714 0.8226 -20.3925 59.5354  False
 中間_近い  近い_近い      7.5 0.9998 -36.5209 51.5209  False
 中間_近い  近い_遠い   20.875  0.892  -26.411  68.161  False
 中間_近い  遠い_中間     -4.0    1.0 -56.2767 48.2767  False
 中間_近い  遠い_近い  20.4545 0.8409 -22.3173 63.2264  False
 中間_近い  遠い_遠い  17.3077 0.9117 -23.4678 58.0832  False
 中間_遠い  近い_中間  18.7381 0.9694 -34.5472 72.0234  False
 中間_遠い  近い_近い   6.6667    1.0 -49.7252 63.0585  False
 中間_遠い  近い_遠い  20.0417 0.9751 -38.9344 79.0177  False
 中間_遠い  遠い_中間  -4.8333    1.0 -67.8813 58.2147  False
 中間_遠い  遠い_近い  19.6212 0.9681 -35.8011 75.0435  False
 中間_遠い  遠い_遠い  16.4744 0.9872 -37.4223  70.371  False
 近い_中間  近い_近い -12.0714 0.9948 -57.2855 33.1426  False
 近い_中間  近い_遠い   1.3036    1.0 -47.0952 49.7023  False
 近い_中間  遠い_中間 -23.5714 0.8908 -76.8567 29.7139  False
 近い_中間  遠い_近い   0.8831    1.0 -43.1158  44.882  False
 近い_中間  遠い_遠い  -2.2637    1.0 -44.3246 39.7971  False
 近い_近い  近い_遠い   13.375 0.9958 -38.4242 65.1742  False
 近い_近い  遠い_中間    -11.5 0.9992 -67.8919 44.8919  False
 近い_近い  遠い_近い  12.9545 0.9941 -34.7594 60.6685  False
 近い_近い  遠い_遠い   9.8077 0.9989 -36.1253 55.7406  False
 近い_遠い  遠い_中間  -24.875 0.9146  -83.851  34.101  False
 近い_遠い  遠い_近い  -0.4205    1.0 -51.1624 50.3215  False
 近い_遠い  遠い_遠い  -3.5673    1.0 -52.6383 45.5037  False
 遠い_中間  遠い_近い  24.4545 0.8922 -30.9678 79.8768  False
 遠い_中間  遠い_遠い  21.3077 0.9398 -32.5889 75.2043  False
 遠い_近い  遠い_遠い  -3.1469    1.0 -47.8841 41.5904  False
-----------------------------------------------------

==================================================

AG - preference グループ平均:
AG_single_segment         近い        中間         遠い
blended_segment                                  
近い                 15.916667 -0.437500  14.500000
中間                  8.600000  7.142857  27.666667
遠い                 13.571429 -3.000000  11.733333

AG - preference 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                         428.485786   2.0  0.308816   
C(AG_single_segment)                      3790.920933   2.0  2.732169   
C(blended_segment):C(AG_single_segment)   1464.135614   4.0  0.527611   
Residual                                 56194.292262  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.735177  
C(AG_single_segment)                     0.071079  
C(blended_segment):C(AG_single_segment)  0.715732  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -4.7589 0.7688 -21.1578   11.64  False
    中間     遠い  -4.0881 0.8283 -20.7401 12.5639  False
    近い     遠い   0.6708 0.9946 -15.4339 16.7755  False
-----------------------------------------------------

Tukey's HSD test for AG_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   11.593 0.1734  -3.7077 26.8936  False
    中間     遠い  15.3871 0.0742   -1.175 31.9492  False
    近い     遠い   3.7941  0.843 -12.4385 20.0268  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い   1.4571    1.0 -36.9693 39.8836  False
 中間_中間  中間_遠い  20.5238 0.8944 -26.1809 67.2285  False
 中間_中間  近い_中間  -7.5804 0.9993 -45.6228 30.4621  False
 中間_中間  近い_近い   8.7738 0.9987 -31.1517 48.6994  False
 中間_中間  近い_遠い   7.3571    1.0 -45.2605 59.9748  False
 中間_中間  遠い_中間 -10.1429 0.9979 -53.5904 33.3047  False
 中間_中間  遠い_近い   6.4286 0.9999 -38.4439  51.301  False
 中間_中間  遠い_遠い   4.5905    1.0  -33.836  43.017  False
 中間_近い  中間_遠い  19.0667  0.853 -21.4844 59.6177  False
 中間_近い  近い_中間  -9.0375 0.9888 -39.2085 21.1335  False
 中間_近い  近い_近い   7.3167 0.9984 -25.1965 39.8298  False
 中間_近い  近い_遠い      5.9    1.0 -41.3406 53.1406  False
 中間_近い  遠い_中間    -11.6 0.9843 -48.3525 25.1525  False
 中間_近い  遠い_近い   4.9714    1.0  -33.455 43.3979  False
 中間_近い  遠い_遠い   3.1333    1.0 -27.5204 33.7871  False
 中間_遠い  近い_中間 -28.1042 0.3975 -68.2915 12.0832  False
 中間_遠い  近い_近い   -11.75 0.9928 -53.7243 30.2243  False
 中間_遠い  近い_遠い -13.1667 0.9973 -67.3553  41.022  False
 中間_遠い  遠い_中間 -30.6667 0.4439 -76.0041 14.6708  False
 中間_遠い  遠い_近い -14.0952 0.9882    -60.8 32.6095  False
 中間_遠い  遠い_遠い -15.9333 0.9418 -56.4844 24.6177  False
 近い_中間  近い_近い  16.3542 0.7878 -15.7043 48.4126  False
 近い_中間  近い_遠い  14.9375 0.9834 -31.9912 61.8662  False
 近い_中間  遠い_中間  -2.5625    1.0 -38.9133 33.7883  False
 近い_中間  遠い_近い  14.0089 0.9598 -24.0335 52.0514  False
 近い_中間  遠い_遠い  12.1708 0.9327 -18.0001 42.3418  False
 近い_近い  近い_遠い  -1.4167    1.0 -49.8845 47.0511  False
 近い_近い  遠い_中間 -18.9167 0.8161 -57.2338 19.4005  False
 近い_近い  遠い_近い  -2.3452    1.0 -42.2708 37.5803  False
 近い_近い  遠い_遠い  -4.1833    1.0 -36.6965 28.3298  False
 近い_遠い  遠い_中間    -17.5 0.9748 -68.9079 33.9079  False
 近い_遠い  遠い_近い  -0.9286    1.0 -53.5462  51.689  False
 近い_遠い  遠い_遠い  -2.7667    1.0 -50.0072 44.4739  False
 遠い_中間  遠い_近い  16.5714 0.9508 -26.8761  60.019  False
 遠い_中間  遠い_遠い  14.7333  0.935 -22.0192 51.4859  False
 遠い_近い  遠い_遠い  -1.8381    1.0 -40.2646 36.5884  False
-----------------------------------------------------

==================================================

AG - believability グループ平均:
AG_single_segment         近い         中間         遠い
blended_segment                                   
近い                 20.750000   4.875000  26.750000
中間                 13.266667  14.285714  24.666667
遠い                 14.428571  -2.625000   6.200000

AG - believability 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                        2051.965989   2.0  1.825976   
C(AG_single_segment)                      2536.196268   2.0  2.256877   
C(blended_segment):C(AG_single_segment)   1661.575160   4.0  0.739290   
Residual                                 45512.434524  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.167628  
C(AG_single_segment)                     0.111221  
C(blended_segment):C(AG_single_segment)  0.567917  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -2.4018 0.9203 -17.1513 12.3478  False
    中間     遠い -10.1976 0.2412 -25.1748  4.7796  False
    近い     遠い  -7.7958 0.4084 -22.2808  6.6891  False
-----------------------------------------------------

Tukey's HSD test for AG_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  11.0825 0.1517  -3.0028 25.1679  False
    中間     遠い   8.8555 0.3532  -6.3912 24.1021  False
    近い     遠い  -2.2271 0.9328 -17.1704 12.7163  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い   -1.019    1.0  -35.601 33.5629  False
 中間_中間  中間_遠い   10.381 0.9969  -31.651 52.4129  False
 中間_中間  近い_中間  -9.4107 0.9936 -43.6471 24.8257  False
 中間_中間  近い_近い   6.4643 0.9997 -29.4668 42.3953  False
 中間_中間  近い_遠い  12.4643 0.9953  -34.889 59.8176  False
 中間_中間  遠い_中間 -16.9107 0.9027 -56.0114   22.19  False
 中間_中間  遠い_近い   0.1429    1.0 -40.2402 40.5259  False
 中間_中間  遠い_遠い  -8.0857 0.9979 -42.6677 26.4962  False
 中間_近い  中間_遠い     11.4 0.9853  -25.094  47.894  False
 中間_近い  近い_中間  -8.3917 0.9863 -35.5441 18.7607  False
 中間_近い  近い_近い   7.4833 0.9961 -21.7769 36.7436  False
 中間_近い  近い_遠い  13.4833 0.9838 -29.0309 55.9975  False
 中間_近い  遠い_中間 -15.8917 0.8374 -48.9671 17.1838  False
 中間_近い  遠い_近い   1.1619    1.0   -33.42 35.7439  False
 中間_近い  遠い_遠い  -7.0667 0.9961 -34.6535 20.5202  False
 中間_遠い  近い_中間 -19.7917 0.7176 -55.9583  16.375  False
 中間_遠い  近い_近い  -3.9167    1.0 -41.6915 33.8582  False
 中間_遠い  近い_遠い   2.0833    1.0 -46.6838 50.8505  False
 中間_遠い  遠い_中間 -27.2917 0.4596 -68.0932 13.5098  False
 中間_遠い  遠い_近い -10.2381 0.9972 -52.2701 31.7939  False
 中間_遠い  遠い_遠い -18.4667 0.7951 -54.9606 18.0273  False
 近い_中間  近い_近い   15.875 0.7116  -12.976  44.726  False
 近い_中間  近い_遠い   21.875 0.7737 -20.3586 64.1086  False
 近い_中間  遠い_中間     -7.5 0.9982  -40.214  25.214  False
 近い_中間  遠い_近い   9.5536  0.993 -24.6828 43.7899  False
 近い_中間  遠い_遠い    1.325    1.0 -25.8274 28.4774  False
 近い_近い  近い_遠い      6.0    1.0 -37.6186 49.6186  False
 近い_近い  遠い_中間  -23.375  0.441 -57.8586 11.1086  False
 近い_近い  遠い_近い  -6.3214 0.9997 -42.2525 29.6096  False
 近い_近い  遠い_遠い   -14.55 0.8101 -43.8103 14.7103  False
 近い_遠い  遠い_中間  -29.375 0.5319 -75.6396 16.8896  False
 近い_遠い  遠い_近い -12.3214 0.9956 -59.6747 35.0319  False
 近い_遠い  遠い_遠い   -20.55 0.8329 -63.0642 21.9642  False
 遠い_中間  遠い_近い  17.0536 0.8984 -22.0471 56.1543  False
 遠い_中間  遠い_遠い    8.825 0.9948 -24.2505 41.9005  False
 遠い_近い  遠い_遠い  -8.2286 0.9976 -42.8105 26.3534  False
-----------------------------------------------------

==================================================

AG - overall_liking グループ平均:
AG_single_segment         近い         中間         遠い
blended_segment                                   
近い                 18.000000   2.125000  16.500000
中間                  9.866667  11.857143  28.666667
遠い                  4.857143  -4.250000   8.600000

AG - overall_liking 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                        2211.689771   2.0  1.409393   
C(AG_single_segment)                      3004.895325   2.0  1.914861   
C(blended_segment):C(AG_single_segment)   1362.952294   4.0  0.434269   
Residual                                 63554.630952  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.250218  
C(AG_single_segment)                     0.153974  
C(blended_segment):C(AG_single_segment)  0.783483  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -4.5179 0.8069 -21.7589 12.7231  False
    中間     遠い -10.0929 0.3586    -27.6  7.4143  False
    近い     遠い   -5.575 0.7131 -22.5067 11.3567  False
-----------------------------------------------------

Tukey's HSD test for AG_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   9.0285 0.3941    -7.42  25.477  False
    中間     遠い  12.0026  0.248  -5.8021 29.8072  False
    近い     遠い   2.9741 0.9131 -14.4763 20.4246  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い  -1.9905    1.0 -42.8561 38.8751  False
 中間_中間  中間_遠い  16.8095 0.9757 -32.8598 66.4789  False
 中間_中間  近い_中間  -9.7321 0.9975 -50.1894 30.7251  False
 中間_中間  近い_近い   6.1429 0.9999  -36.317 48.6027  False
 中間_中間  近い_遠い   4.6429    1.0 -51.3147 60.6004  False
 中間_中間  遠い_中間 -16.1071 0.9709 -62.3126 30.0983  False
 中間_中間  遠い_近い     -7.0 0.9999 -54.7208 40.7208  False
 中間_中間  遠い_遠い  -3.2571    1.0 -44.1228 37.6085  False
 中間_近い  中間_遠い     18.8 0.8986 -24.3251 61.9251  False
 中間_近い  近い_中間  -7.7417 0.9974 -39.8277 24.3444  False
 中間_近い  近い_近い   8.1333 0.9978 -26.4436 42.7103  False
 中間_近い  近い_遠い   6.6333    1.0 -43.6058 56.8725  False
 中間_近い  遠い_中間 -14.1167 0.9641 -53.2021 24.9687  False
 中間_近い  遠い_近い  -5.0095    1.0 -45.8751 35.8561  False
 中間_近い  遠い_遠い  -1.2667    1.0 -33.8661 31.3328  False
 中間_遠い  近い_中間 -26.5417  0.562 -69.2799 16.1966  False
 中間_遠い  近い_近い -10.6667 0.9976 -55.3053  33.972  False
 中間_遠い  近い_遠い -12.1667  0.999  -69.795 45.4616  False
 中間_遠い  遠い_中間 -32.9167  0.431  -81.132 15.2986  False
 中間_遠い  遠い_近い -23.8095 0.8391 -73.4789 25.8598  False
 中間_遠い  遠い_遠い -20.0667 0.8601 -63.1917 23.0584  False
 近い_中間  近い_近い   15.875 0.8596 -18.2184 49.9684  False
 近い_中間  近い_遠い   14.375 0.9913 -35.5326 64.2826  False
 近い_中間  遠い_中間   -6.375 0.9998 -45.0332 32.2832  False
 近い_中間  遠い_近い   2.7321    1.0 -37.7251 43.1894  False
 近い_中間  遠い_遠い    6.475 0.9993 -25.6111 38.5611  False
 近い_近い  近い_遠い     -1.5    1.0 -53.0443 50.0443  False
 近い_近い  遠い_中間   -22.25   0.72 -62.9994 18.4994  False
 近い_近い  遠い_近い -13.1429 0.9861 -55.6027  29.317  False
 近い_近い  遠い_遠い     -9.4 0.9941  -43.977  25.177  False
 近い_遠い  遠い_中間   -20.75 0.9521  -75.421  33.921  False
 近い_遠い  遠い_近い -11.6429 0.9991 -67.6004 44.3147  False
 近い_遠い  遠い_遠い     -7.9 0.9999 -58.1392 42.3392  False
 遠い_中間  遠い_近い   9.1071 0.9994 -37.0983 55.3126  False
 遠い_中間  遠い_遠い    12.85 0.9797 -26.2354 51.9354  False
 遠い_近い  遠い_遠い   3.7429    1.0 -37.1228 44.6085  False
-----------------------------------------------------

==================================================

AG - persuasiveness グループ平均:
AG_single_segment         近い        中間         遠い
blended_segment                                  
近い                 12.083333 -1.187500  26.000000
中間                  6.666667  8.142857  22.833333
遠い                  3.857143  0.250000   7.600000

AG - persuasiveness 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                        1075.221091   2.0  0.792152   
C(AG_single_segment)                      2922.075661   2.0  2.152793   
C(blended_segment):C(AG_single_segment)   1364.830887   4.0  0.502759   
Residual                                 54972.335119  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.456346  
C(AG_single_segment)                     0.122760  
C(blended_segment):C(AG_single_segment)  0.733772  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -3.3125  0.876 -19.4165 12.7915  False
    中間     遠い  -5.7333 0.6818 -22.0859 10.6193  False
    近い     遠い  -2.4208 0.9293 -18.2359 13.3943  False
-----------------------------------------------------

Tukey's HSD test for AG_single_segment:
Multiple Comparison of Means - Tukey HSD, FWER=0.05 
====================================================
group1 group2 meandiff p-adj   lower   upper  reject
----------------------------------------------------
    中間     近い   6.7097 0.5465 -8.5019 21.9212  False
    中間     遠い  12.9097 0.1537  -3.556 29.3754  False
    近い     遠い      6.2 0.6317 -9.9381 22.3381  False
----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い  -1.4762    1.0 -39.4826 36.5302  False
 中間_中間  中間_遠い  14.6905 0.9835 -31.5037 60.8846  False
 中間_中間  近い_中間  -9.3304 0.9969 -46.9569 28.2962  False
 中間_中間  近い_近い   3.9405    1.0 -35.5486 43.4296  False
 中間_中間  近い_遠い  17.8571 0.9736 -34.1852 69.8995  False
 中間_中間  遠い_中間  -7.8929 0.9996 -50.8654 35.0797  False
 中間_中間  遠い_近い  -4.2857    1.0 -48.6676 40.0962  False
 中間_中間  遠い_遠い  -0.5429    1.0 -38.5492 37.4635  False
 中間_近い  中間_遠い  16.1667  0.933 -23.9411 56.2744  False
 中間_近い  近い_中間  -7.8542 0.9953 -37.6953 21.9869  False
 中間_近い  近い_近い   5.4167 0.9998 -26.7411 37.5744  False
 中間_近い  近い_遠い  19.3333 0.9228 -27.3908 66.0574  False
 中間_近い  遠い_中間  -6.4167 0.9997 -42.7674 29.9341  False
 中間_近い  遠い_近い  -2.8095    1.0 -40.8159 35.1969  False
 中間_近い  遠い_遠い   0.9333    1.0 -29.3853 31.2519  False
 中間_遠い  近い_中間 -24.0208 0.5981 -63.7689 15.7272  False
 中間_遠い  近い_近い   -10.75 0.9958 -52.2655 30.7655  False
 中間_遠い  近い_遠い   3.1667    1.0 -50.4296 56.7629  False
 中間_遠い  遠い_中間 -22.5833 0.7993 -67.4252 22.2585  False
 中間_遠い  遠い_近い -18.9762 0.9258 -65.1703 27.2179  False
 中間_遠い  遠い_遠い -15.2333 0.9519 -55.3411 24.8744  False
 近い_中間  近い_近い  13.2708  0.918 -18.4371 44.9788  False
 近い_中間  近い_遠い  27.1875 0.6379 -19.2282 73.6032  False
 近い_中間  遠い_中間   1.4375    1.0 -34.5159 37.3909  False
 近い_中間  遠い_近い   5.0446    1.0 -32.5819 42.6712  False
 近い_中間  遠い_遠い   8.7875   0.99 -21.0536 38.6286  False
 近い_近い  近い_遠い  13.9167 0.9909 -34.0213 61.8546  False
 近い_近い  遠い_中間 -11.8333 0.9853 -49.7316 26.0649  False
 近い_近い  遠い_近い  -8.2262 0.9991 -47.7153 31.2629  False
 近い_近い  遠い_遠い  -4.4833    1.0 -36.6411 27.6744  False
 近い_遠い  遠い_中間   -25.75 0.7944 -76.5958 25.0958  False
 近い_遠い  遠い_近い -22.1429 0.9106 -74.1852 29.8995  False
 近い_遠い  遠い_遠い    -18.4 0.9411 -65.1241 28.3241  False
 遠い_中間  遠い_近い   3.6071    1.0 -39.3654 46.5797  False
 遠い_中間  遠い_遠い     7.35 0.9993 -29.0007 43.7007  False
 遠い_近い  遠い_遠い   3.7429    1.0 -34.2635 41.7492  False
-----------------------------------------------------

==================================================

AG - interest グループ平均:
AG_single_segment         近い        中間         遠い
blended_segment                                  
近い                 14.750000  0.812500  16.250000
中間                  6.200000  5.714286  30.000000
遠い                  5.428571 -3.000000   8.666667

AG - interest 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                        1322.106044   2.0  0.913956   
C(AG_single_segment)                      3542.228616   2.0  2.448699   
C(blended_segment):C(AG_single_segment)   1523.333586   4.0  0.526531   
Residual                                 58586.313690  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.405029  
C(AG_single_segment)                     0.092780  
C(blended_segment):C(AG_single_segment)  0.716515  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -3.2098 0.8907    -19.9 13.4804  False
    中間     遠い  -6.3786 0.6434 -23.3264 10.5693  False
    近い     遠い  -3.1688 0.8896 -19.5595  13.222  False
-----------------------------------------------------

Tukey's HSD test for AG_single_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い   8.1233 0.4382  -7.6117 23.8583  False
    中間     遠い  14.0645  0.126  -2.9678 31.0968  False
    近い     遠い   5.9412 0.6739 -10.7523 22.6347  False
-----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い   0.4857    1.0 -38.7501 39.7215  False
 中間_中間  中間_遠い  24.2857 0.7894 -23.4027 71.9741  False
 中間_中間  近い_中間  -4.9018    1.0 -43.7455 33.9419  False
 中間_中間  近い_近い   9.0357 0.9986 -31.7307 49.8022  False
 中間_中間  近い_遠い  10.5357 0.9994 -43.1901 64.2615  False
 中間_中間  遠い_中間  -8.7143 0.9994 -53.0769 35.6484  False
 中間_中間  遠い_近い  -0.2857    1.0 -46.1033 45.5318  False
 中間_中間  遠い_遠い   2.9524    1.0 -36.2834 42.1882  False
 中間_近い  中間_遠い     23.8 0.6611 -17.6051 65.2051  False
 中間_近い  近い_中間  -5.3875 0.9997 -36.1939 25.4189  False
 中間_近い  近い_近い     8.55 0.9959  -24.648  41.748  False
 中間_近い  近い_遠い    10.05 0.9991 -38.1855 58.2855  False
 中間_近い  遠い_中間     -9.2 0.9971 -46.7266 28.3266  False
 中間_近い  遠い_近い  -0.7714    1.0 -40.0072 38.4644  False
 中間_近い  遠い_遠い   2.4667    1.0 -28.8327  33.766  False
 中間_遠い  近い_中間 -29.1875 0.3741 -70.2213 11.8463  False
 中間_遠い  近い_近い   -15.25 0.9671 -58.1084 27.6084  False
 中間_遠い  近い_遠い   -13.75 0.9968 -69.0799 41.5799  False
 中間_遠い  遠い_中間    -33.0 0.3711 -79.2923 13.2923  False
 中間_遠い  遠い_近い -24.5714 0.7786 -72.2598  23.117  False
 中間_遠い  遠い_遠い -21.3333 0.7787 -62.7385 20.0718  False
 近い_中間  近い_近い  13.9375 0.9102 -18.7961 46.6711  False
 近い_中間  近い_遠い  15.4375 0.9821 -32.4796 63.3546  False
 近い_中間  遠い_中間  -3.8125    1.0  -40.929  33.304  False
 近い_中間  遠い_近い   4.6161    1.0 -34.2276 43.4598  False
 近い_中間  遠い_遠い   7.8542 0.9962 -22.9522 38.6606  False
 近い_近い  近い_遠い      1.5    1.0 -47.9886 50.9886  False
 近い_近い  遠い_中間   -17.75 0.8762 -56.8742 21.3742  False
 近い_近い  遠い_近い  -9.3214 0.9982 -50.0879  31.445  False
 近い_近い  遠い_遠い  -6.0833 0.9996 -39.2813 27.1146  False
 近い_遠い  遠い_中間   -19.25 0.9607 -71.7406 33.2406  False
 近い_遠い  遠い_近い -10.8214 0.9993 -64.5473 42.9044  False
 近い_遠い  遠い_遠い  -7.5833 0.9999 -55.8189 40.6522  False
 遠い_中間  遠い_近い   8.4286 0.9995 -35.9341 52.7912  False
 遠い_中間  遠い_遠い  11.6667 0.9857 -25.8599 49.1933  False
 遠い_近い  遠い_遠い   3.2381    1.0 -35.9977 42.4739  False
-----------------------------------------------------

==================================================

AG - click_likelihood グループ平均:
AG_single_segment         近い        中間         遠い
blended_segment                                  
近い                 16.916667 -0.812500  18.500000
中間                 13.200000  6.285714  30.666667
遠い                  5.000000 -9.375000  13.866667

AG - click_likelihood 二元配置分散分析結果:
                                               sum_sq    df         F  \
C(blended_segment)                        2173.221732   2.0  1.347180   
C(AG_single_segment)                      6912.886454   2.0  4.285298   
C(blended_segment):C(AG_single_segment)    595.903427   4.0  0.184700   
Residual                                 65333.124405  81.0       NaN   

                                           PR(>F)  
C(blended_segment)                       0.265734  
C(AG_single_segment)                     0.017019  
C(blended_segment):C(AG_single_segment)  0.945734  
Residual                                      NaN  

Tukey's HSD test for blended_segment:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
    中間     近い  -6.9643 0.6228 -24.8187 10.8902  False
    中間     遠い  -9.6143  0.419 -27.7443  8.5158  False
    近い     遠い    -2.65  0.931 -20.1842 14.8842  False
-----------------------------------------------------

Tukey's HSD test for AG_single_segment:
Multiple Comparison of Means - Tukey HSD, FWER=0.05 
====================================================
group1 group2 meandiff p-adj   lower   upper  reject
----------------------------------------------------
    中間     近い  14.2429 0.1065 -2.3244 30.8102  False
    中間     遠い  20.0594 0.0245  2.1262 37.9926   True
    近い     遠い   5.8165 0.7107  -11.76 23.3929  False
----------------------------------------------------

Tukey's HSD test for interaction groups:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
group1 group2 meandiff p-adj   lower    upper  reject
-----------------------------------------------------
 中間_中間  中間_近い   6.9143 0.9998 -34.5192 48.3477  False
 中間_中間  中間_遠い   24.381 0.8316 -25.9786 74.7405  False
 中間_中間  近い_中間  -7.0982 0.9998 -48.1176 33.9212  False
 中間_中間  近い_近い   10.631 0.9969 -32.4189 53.6808  False
 中間_中間  近い_遠い  12.2143 0.9988 -44.5208 68.9494  False
 中間_中間  遠い_中間 -15.6607 0.9775 -62.5082 31.1867  False
 中間_中間  遠い_近い  -1.2857    1.0 -49.6696 47.0981  False
 中間_中間  遠い_遠い    7.581 0.9996 -33.8525 49.0144  False
 中間_近い  中間_遠い  17.4667 0.9362 -26.2576  61.191  False
 中間_近い  近い_中間 -14.0125 0.9047 -46.5444 18.5194  False
 中間_近い  近い_近い   3.7167    1.0 -31.3408 38.7741  False
 中間_近い  近い_遠い      5.3    1.0 -45.6373 56.2373  False
 中間_近い  遠い_中間  -22.575 0.6718 -62.2035 17.0535  False
 中間_近い  遠い_近い     -8.2 0.9994 -49.6335 33.2335  False
 中間_近い  遠い_遠い   0.6667    1.0 -32.3858 33.7191  False
 中間_遠い  近い_中間 -31.4792 0.3455 -74.8113  11.853  False
 中間_遠い  近い_近い   -13.75 0.9877 -59.0089 31.5089  False
 中間_遠い  近い_遠い -12.1667 0.9991 -70.5957 46.2624  False
 中間_遠い  遠い_中間 -40.0417 0.1986 -88.9269  8.8436  False
 中間_遠い  遠い_近い -25.6667 0.7886 -76.0262 24.6928  False
 中間_遠い  遠い_遠い    -16.8 0.9487 -60.5243 26.9243  False
 近い_中間  近い_近い  17.7292 0.7829 -16.8379 52.2963  False
 近い_中間  近い_遠い  19.3125 0.9506 -31.2885 69.9135  False
 近い_中間  遠い_中間  -8.5625 0.9987 -47.7579 30.6329  False
 近い_中間  遠い_近い   5.8125 0.9999 -35.2069 46.8319  False
 近い_中間  遠い_遠い  14.6792 0.8795 -17.8528 47.2111  False
 近い_近い  近い_遠い   1.5833    1.0 -50.6772 53.8439  False
 近い_近い  遠い_中間 -26.2917 0.5288 -67.6072 15.0239  False
 近い_近い  遠い_近い -11.9167 0.9933 -54.9665 31.1332  False
 近い_近い  遠い_遠い    -3.05    1.0 -38.1074 32.0074  False
 近い_遠い  遠い_中間  -27.875 0.8005 -83.3057 27.5557  False
 近い_遠い  遠い_近い    -13.5 0.9976 -70.2351 43.2351  False
 近い_遠い  遠い_遠い  -4.6333    1.0 -55.5706 46.3039  False
 遠い_中間  遠い_近い   14.375 0.9869 -32.4725 61.2225  False
 遠い_中間  遠い_遠い  23.2417 0.6364 -16.3868 62.8702  False
 遠い_近い  遠い_遠い   8.8667 0.9989 -32.5668 50.3001  False
-----------------------------------------------------

==================================================

linear regression

Features to be regressed:

1. [y] each of the constructs(same questions for 4 different(big5) personalized contents) vs [X] big5 distances, controlling for socioeconomic
2. [y] 'perceived_credibility' scores for each of 4 versions vs [X] 'perceived_personalization' scores for each of 4 versions & matched big5 scores for each of 4 versions, controlling for socioeconomic
3. [y] 'overall_attitude' scores for each of 4 versions vs [X] 'perceived_personalization' scores for each of 4 versions & (moderator) 'perceived_credibility' scores for each of 4 versions & matched big5 scores for each of 4 versions, controlling for socioeconomic
4. [y] 'ads_engagement' scores for each of 4 versions vs [X] 'overall_attitude' scores for each of 4 versions & matched big5 scores for each of 4 versions, controlling for socioeconomic